Building websites for
return on investment
written by PAUL BOAG
This book accompanies season one of the Boagworld Show
This book is dedicated to Matthew Curry
for teaching me that making money from
websites does not make you evil.
How to ensure your website generates a
return on investment
Written by Paul Boag
Edited by Relly Annett-Baker
Why do you have a website?
It is such a straightforward question that it seems almost
absurd to ask but you might be amazed how many
website owners can’t answer it at all. Even when they do
have a response it is often vague and unfocussed.
Common answers include:
“Our competition have one.”
“We need an ‘online presence’.”
“Doesn’t everybody have a website these days?”
The reality is that few of us question what we gain from
having a website. This shortsighted view that ‘we have a
website because, well, we have to’ portrays the website
as an expense that has to be paid and yet provides no
easily articulated benefits. In many ways it feels like an
arbitrary tax. You know that the tax must be paid because
it pays for services and whatnot, however, because it is
hard to articulate what exactly the benefit is to you, you
can end up resenting paying the tax.
When an organisation perceives its website as a costly
burden that provides no tangible benefit, their site is
never going to receive the investment it requires. This not
only puts budgets for further investment in jeopardy, it
also inevitably leads to management questioning the
need to have a web team working on the site. For
instance, perhaps your role?
If you are keen to see ongoing investment in your
website, not to mention job security for yourself, the time
has come to ask: What return on investment (ROI) does
your site offer?
Does your website save your organisation money?
Does it generate income through ecommerce? Does it
provide leads for your sales team to follow up? In short,
does it create a tangible, trackable benefit that can be
presented to management as evidence that your site is
worth investing in? Ultimately, your senior management
want to understand what they get back from investing
money and energy into their online presence. It is down
to you to explain what those returns are for the business.
This book will help you do just that. We begin by
establishing the business objectives for your site. We
refine these objectives into specific, measurable goals.
We then look at improving your site in order to meet
those goals through a cycle of development, testing and
tracking. By the end of this book you will have a clear
idea of how to ensure your website provides tangible
returns for your business and how best to present those
to other people in the business so they understand how
valuable your site, and you, really are.
CHAPTER ONE (7)
Using business objectives to inform decisions
CHAPTER TWO (23)
Measuring your successes and failures
CHAPTER THREE (37)
Becoming user focused
CHAPTER FOUR (67)
Creating your calls to action
CHAPTER FIVE (95)
A cycle of refinement
CHAPTER SIX (123)
I have a confession to make: I am addicted to buying
gadgets. If it is shiny and new, and especially if it has an
Apple logo, I want it. Rarely do I question my desire, I just
find myself buying it. Even when I do ask myself why I
should need this shiny and new thing, I convince myself
with a few weak justifications like “This will be what finally
boosts my productivity to 110%” or “I have to keep up
with technology for work, don’t I?” before handing over
my credit card.
Perhaps you judge me for my unchecked spending,or
how easily I am swayed by the shiny, shiny gadgetry?
However, just stop a moment and ask yourself: do you
carefully consider every expenditure on your site or are
you sometimes tempted by the idea of a shiny, new
design to fix all your problems? If so, you are not alone.
It is not enough to simply ‘want’ a new feature or
design. You need to ask whether it provides a benefit to
the business. There is one question that, if you ask it over
and over, will become a mantra for developing your site
for greater returns.
“Does this help my organisation fulfil our business
objectives for the site?”
1. Why care about business objectives?
2. How to identify business objectives
3. Macro and micro objectives
4. Be specific
5. Avoid unrealistic objectives
6. Avoiding the blame game
7. Next Actions
Use business objectives to inform decisions 9
Why care about business objectives?
In our haste to get a shiny, new website it is easy to
overlook business objectives. They are often considered a
‘nice to have’, an optional part of the planning process, or
perhaps a bonus presuming time and budget allow us to
think about such things. However, ignoring business
objectives is a false economy. For a start, they are an
invaluable tool for making decisions.
Business objectives help decision making
Business Business objectives act as a plumb line which a project
objectives act can be measured against, for an unbiased evaluation.
as a plumb While working on any web project there are countless
line which a decisions to be made such as:
project can ‣ Functionality
be measured ‣ User testing
against, for ‣ Time spent on design
an unbiased ‣ Accessibility
evaluation. ‣ Maintainability
…and many more.
Business objectives allow us to make informed
decisions based on what return these individual aspects,
and expenses, will bring.
For example, if one of your business objectives is
‘Reduce cart abandonment at checkout’, you can be sure
that investing in usability testing will be a valuable way to
find the cause.
10 Use business objectives to inform decisions
Equally, we know that forcing users to register, an idea
often suggested by over enthusiastic marketing
departments, is a bad idea because users hate the
process and it will only increase cart abandonment.
Investment in the user experience design and copy will
pay rich dividends here.
Business objectives justify investment
Business objectives help us judge the quality of an idea
and justify the expense of implementation. This is
particularly important when dealing with senior
management. They can be used both to prove further
investment is justified and to defend money already spent
on the site.
However, decision making is not the only benefit.
Business objectives are also a communication tool.
Business objectives help senior management decide
whether to invest further in a website.
Use business objectives to inform decisions 11
Business objectives aid communication
Having clearly Many web projects fail because of a lack of
defined communication between the different parties involved.
business I've witnessed multiple projects where the developers’
objectives expectations were radically different from that of
helps to management. Often these differences are not discovered
until the end of the project and this inevitably leads to
conflict. Having clearly defined business objectives helps
to ensure all parties are working towards the same
same I'm not claiming that this is the solution to all
communication problems, as different parties will
interpret business objectives in different ways. However,
working towards a common aim is a starting point for
Now the benefits of business objectives are clear, the
next question is: “How can I identify what my business
objectives should be?”
How to identify business objectives
In many cases, business objectives are pretty obvious. If
we run an ecommerce site together, increased sales is an
obvious objective. If we run a web design company, the
primary objective is probably lead generation.
However, on some sites the objectives are not so clear.
For example, what is the objective of a museum website?
What about a news website?
12 Use business objectives to inform decisions
Every site has business objectives
The truth is that all websites have business objectives. It is
just that some are not as immediately obvious as our
ecommerce site or web design company.
Take our museum website. The primary purpose of this
website is probably to encourage more people to visit the
museum but, like most websites, I suspect our museum
has more than a single objective. The museum might
want people to sign up to their newsletter or they may
provide resources that research students will find useful to
encourage them to join the museum staff. Just because it
is not immediately obvious how you track these
objectives does not make them any less important in
defining where resources should be spent. In Chapter 2,
(Measuring your success) we will explore how to track
these less tangible goals.
Nebulous business objectives such as customer
satisfaction or improved perception of your brand, are
just as important as the measurable ones such as
purchases, signups or lead generation.
However, its not just the type of business objectives
that matters. It is also how you decide on them.
Work collaboratively to set objectives
At this point we might be feeling smug because we can
list our business objectives. It is something that we have
put a lot of thought into and so we feel ahead of the
However, could all of our colleagues who have a
vested interest in the website (our stakeholders) also
repeat the business objectives? Do they actually know
what they are? Even more importantly were they involved
in creating them?
Use business objectives to inform decisions 13
At ﬁrst glance you may wonder what business objectives a charity like Butterﬂy
Conservation could have beyond raising donations. However, they realised that the web
could be utilised to mobilise thousands of people to help track butterﬂy populations
across the UK. This led to a site focused exclusively on this objective.
14 Use business objectives to inform decisions
Business objectives cannot be created in isolation. It
must be a collaborative process with everybody agreeing
on the final outcome.
A few paragraphs ago, I mentioned how business For business
objectives can be used to assess ideas proposed by objectives to
various people internally within the organisation. Over at be most
our ecommerce business, our marketing department were effective they
proposing that users were forced to register before need to be as
purchasing so they could get the skinny on our
customers. I suggested that if one of the business
objectives was to increase orders, you could argue that
this idea would be go against that objective as it
increased instances of cart abandonment. Where this falls
down is when the marketing department are not involved
in agreeing the business objectives. If that is the case they
are not going to accept them as a legitimate reason for
rejecting their idea.
That is why it is important to consult widely about
what your business objectives should be.
How you manage this process is entirely up to you.
However, I would give one piece of advice: Do not allow
your list of business objectives to grow too long and
By keeping the list short you reduce the possibility of
different objectives clashing with one another. Also, by
prioritising the list, you ensure that when a clash does
happen the most important objective is obvious. It is also
important to remember that different objectives can
apply to different parts of the site. This is when you need
to start thinking in terms of macro and micro objectives.
Use business objectives to inform decisions 15
Macro and micro objectives
Depending on the size of your site and organisation it
may be necessary to think of your site as a collection of
smaller sites all with their own business objectives.
Take a University website. These sites are traditionally
extremely large and consist of many different sub-sites
trying to accommodate diverse audiences. In such a
situation, it is very difficult to expect the postgraduate
Physics departmental research sub-site to share the same
business objectives as undergraduate student
Website such as the University of Northampton have
such a broad audience that it requires both macro and
In such a situation, I recommend having two levels of
The macro objectives are those that apply across the
whole site. These define the priorities for top level pages
and especially the homepage. This goes someway to
defusing homepage and navigational disputes.
16 Use business objectives to inform decisions
The micro objectives are those that apply within an
individual sub-site. So where as the macro objectives may
place undergraduate student recruitment before
enhancing the University’s research status, this would not
hold true for the research sub-site. In this case, they
would have their own specific micro objectives, for
example, they might want more subscriptions to the
research departments newsletter.
Notice that I mention these micro objectives should be
specific. That does not mean your macro objectives can
afford to be vague. All objectives should be as specific as
One problem that occurs when writing business
objectives is that they often become vague, especially
when produced as part of a collaborative process. For
business objectives to be most effective they need to be
as specific as possible.
Take our news website I mentioned earlier. I suggested
our goal might be to increase advertising revenue.
However, this could be much more specific by identifying
how the revenue could be increased.
Instead of the vague objective of “increasing
revenue”, why not have two goals?:
‣ Generate more traffic to the website
‣ Increase the number of users clicking on
The same is true for our ecommerce website. Instead
of simply looking to “increase sales”, we can break the
objective down into:
‣ Increase the number of sales
‣ Increase the average order value
Use business objectives to inform decisions 17
One reason it is important to be as specific as possible
is that otherwise you can get undesirable results. For
example, if you just set your business objective as
“Generating more leads” (as we discussed with our
fictional web design agency) it could end up costing the
company money as they frantically follow up a large
number of low quality leads that will almost certainly
never go anywhere. Instead, the business objective
should be to generate more high quality leads.
Make it measurable
Where ever possible attach a measurable value to your
business objective. This makes it easier to judge whether
the objective has generated a sufficient return on
For example, let’s say you wanted to increase
subscriptions to your newsletter via your website. You
may decide that you would be willing to pay £1 for each
new subscriber. That figure may be arbitrary or based on
a knowledge of how many subscribers convert to
Whichever the case, once you have decided on how
much you are willing to pay for a subscriber it allows you
to calculate how many more subscribers you require to
generate a return on investment.
Let me explain what I mean: If it costs £5000 to
change the website in a way that highlights the
newsletter, you know that you have to generate at least
5000 additional subscribers to make the investment
worthwhile. Therefore, our business objective might be to
generate 5000 additional newsletter subscribers in the
first 6 months.
18 Use business objectives to inform decisions
Before you write the cheque for the next piece of
development work, decide what kind of return you expect
for the money.
The advantage of this quantitative approach is that
success can be easily measured and proved to others,
such as members of senior management. In turn, this
makes future investment easier to obtain as you and the
site have already gained a reputation for delivering
Of course, you will only gain a reputation for
delivering returns if your objectives are reasonable.
Avoid unrealistic objectives
It is remarkably easy for business objectives to become
unreasonable. This is often because they, or rather their
authors, assume too much return for too little investment
or the returns to happen in too short a timescale.
Remember that these objectives need to be agreed by
everybody. This doesn’t just mean the stakeholders who
want something delivered. You also need to include the
team who are actually going to implement the web
Use business objectives to inform decisions 19
project. Whether you are using an internal web team or
outsourcing, you need to make sure that they can deliver
on those objectives.
This is a particular problem with external agencies who
may tell you anything to win the work. A better approach
is to engage an agency before defining exactly what
needs to be built because then they can be involved in
the process of defining realistic objectives.
The problem with having unrealistic goals is not just a
failure to deliver, it is also the inevitable blame game that
Avoiding the blame game
Assigning blame is damaging to the morale of your web
team. It’s also unrealistic to attribute blame to any
individual. Web projects are complex, with each person’s
contribution depending on many others. If you start
blaming developers for late delivery, they’re just as likely
to blame you for a poorly defined scope of work.
Everyone is unhappy. Nobody wins.
A better approach is, when the project is over, to
discuss how it did or didn’t fulfil the business objectives.
Usually this can’t happen immediately because business
objectives are fulfilled over time. When that meeting does
take place, it should always look forward rather than focus
on past mistakes.
Consider which parts of the project didn’t meet
expectations. Was it because the expectations were
unrealistic? Did a problem arise that couldn’t have been
anticipated in advance?
20 Use business objectives to inform decisions
When things didn't go to
plan with the new website,
things got ugly...
It is unrealistic to attribute blame when working on
complex web projects where roles are so interdependent.
Ask these questions to ensure that these issues can be
overcome in the future. Concluding that the business
objectives were unrealistic enables you to revise them
going forward. Identifying that a project was delayed due
to unforeseen problems tells you to build in contingency
plans next time.
Hopefully, I have now convinced you how important
business objectives are.
However, knowledge is one thing and implementation
is another. Before proceeding to the next chapter I
recommend you complete the following actions:
Action 1: List your objectives
The first step is to bring all the stakeholders together
including management and your web team. Draw up an
initial set of business objectives. The aim is to agree a
maximum of half a dozen prioritised objectives.
Action 2: Establish measurable goals
Now is the time to get specific with your objectives.
Identify detailed objectives within your general aims
and attach specific numbers so they are measurable.
Share these measurable goals and iterate until you
have unanimous agreement.
Action 3: Review using these objectives
Use your approved objectives as a plumb line against
which to make decisions. Does your current web
project help achieve your business objectives? Will it
generate the required level of return? Is your current
site doing its job or does it need to change?
Although you will see immediate benefits from having a
set of business objectives, the real returns emerge over
time. Once you can measure the successes and failures of
changes it is possible to track the returns they provide. It
is this measuring of return we are going to address in the
They say opposites attract and none more so than myself
and co-founder of Headscape, Chris Scott. While I think in
broad concepts, he obsesses over detail. While I act on
gut instinct, he carefully examines all of the different
One area where I wish I was naturally like Chris is over
his desire to quantify success. I am great at establishing
objectives whether for our business, website or even in
my personal life. However, I quickly lose interest when it
comes to being specific and tracking my successes and
failures. If Chris undertakes a project he wants to know
exactly what he is trying to achieve and then wants
empirical evidence showing whether or not that objective
has been met. In short, he measures success by cold hard
I think many website owners could learn a lot from
Chris' obsession with data. There is no point defining
business objectives unless we track whether those
objectives are being fulfilled. Too many of us act on gut
instinct and fail to analyse whether we are actually
achieving what we set out to do. When we do decide that
some form of measurement is required, we often become
caught up in the functionality of the tool we are using
rather than what we a trying to measure.
1. Focus on goals not tools
2. Everything is trackable. Well, almost
3. Finding the right tool for the job
4. Tracking your failures
5. Next Actions
Measuring your successes and failures 25
Focus on goals not tools
When you Be honest. When you last opened Google Analytics did
open Google you know what you were looking for? Most of us don't.
Analytics do We might look at page views, unique visitors and dwell
you really time and if the numbers are going up, we are happy.
understand However, what do these figures really tell us? Are they
what you actually that much help?
should be The number of page views or visitors are really only
looking for? useful in the early days of your website, when you are
trying to build an audience, and even then, it is more a
reflection of your marketing skills than the effectiveness
of your website. Unfortunately, most of us never move
beyond these basic figures to examine the data that really
tells us whether we are meeting our business objectives.
Part of the problem is that we let the tool dictate what
we measure. We install Google Analytics because that
'tracks stuff' but don't really consider whether it is the
right tool for the job, or what we need to know to
measure the success of meeting our business objectives.
When we do open the app we tend to look at the default
information it offers at the top level and use that as our
measure of success. Instead, we should be using it to
track our real objectives.
Worse still, I often hear website owners say their
business objective cannot be tracked. Often this is
because they are limited in their thinking to the tools they
currently have. It is certainly true that Google Analytics
cannot track everything. However with a little lateral
thinking and effort to find the right tools, anything can be
26 Measuring your successes and failures
Everything is trackable. Well, almost.
It is fair to say that some business objectives are easier to With a little
track than others. For example, if your objective is to creative
increase sales by 30% over the next quarter, this is fairly thinking it is
easy to track through the reporting tools on your possible to
ecommerce platform. Equally, if you want to increase the track almost
number of people who signup to your newsletter it is any business
relatively easy to build or buy a tool that would allow you
to track those figures.
However, some objectives are not so easy to track.
What if you are looking to improve how your brand is
perceived online or increase levels of customer
satisfaction? These are much harder to track. Although
defining these objectives can still be used as a plumb line
for decision making, they do lose their power if they
cannot be tracked. So what am I saying? Am I suggesting
you should give up attempting to track these kinds of
business objectives? Not at all.I believe that with a bit of
lateral thinking and a willingness to compromise it is
possible to track almost anything.
Let's say you wanted to track the leads generated
through your website. Tracking enquiries submitted
through a contact us form is easy. However, what about
email? What about those that pick up the phone and call
you? With a bit of thought even these can be tracked.
What about creating an email address that is shown
nowhere other than on the website? You can even have a
unique telephone number that would allow you to track
enquiries from the site. You don't even need to be that
advanced. Sometimes it is enough to simply ask people
whether they visited the website. It doesn't need to be
Another example of a more challenging business
objective is perception. How do you measure whether the
website is improving how people perceive your brand?
Measuring your successes and failures 27
Believe it or not, even something as nebulous as
perception is possible to track.
To begin with, there are online web apps that allow
you to track mentions of your brand and whether they
have a positive or negative context. You could also run
design testing to see what words users associate with
your website. Both of these approaches would give you
an indication of whether changes to your site are
improving the way your brand is perceived.
Social Mention is just one of many apps that allow you
to see how people are reacting to your brand.
Finally, what about our museum example I gave in
chapter one? I suggested one of the aims was to
encourage more people to visit the museum. How can
you tell whether new visitors came after seeing the
website? One way would be to offer coupon codes,
discounts or other offers that can be redeemed at the
museum. If you give 20% off the entrance fee if they
mention the website, this would allow you to easily track
who had come through that route.
Of course, none of these methods are not perfect.
They have some obvious holes that could compromise the
results you receive. However, I would argue that some
data is better than none. Better to track what you can
rather than make changes with no way of telling whether
they are an improvement or not. It comes down to
28 Measuring your successes and failures
looking for the best tool available and being creative in
the way you use it.
Finding the right tool for the job
I am amazed at just how many free tools are available to
help track online activity but ultimately you need the
ability to do just two things: collect data and display that
data in some usable format. That doesn't always require a
fancy web app. After all, even a simple spreadsheet can
do that. So can pen and paper.
I have already mentioned stats tools, such as Google
Analytics, and tools for tracking online brand perceptions.
However, this is just the tip of a very big iceberg. There
are tools such as Postrank Analytics which track
engagement with your website through sites like Twitter,
Facebook and Delicious. There are also tools like
Feedburner, which track and manages subscriptions to
email newsletters and RSS feeds. You also have the
option to run polls and surveys on your website for
tracking customer satisfaction.
Tools like Click Tales allows you to build up a detailed
picture of user behaviour.
Measuring your successes and failures 29
There are even tools, such as Get Satisfaction and User
Voice, which allow users to submit improvements they
think could be made to your site or products. It is also
possible to use usage-analysis services like Click Tales to
build up a detailed picture of how users are interacting
with your site. Finally, for those with bigger budgets,
there are numerous Customer Relationship Management
systems (CRM) for recording all kinds of data about those
users with whom you interact.
Get Satisfaction allow users to directly feedback to an
organisation how they feel about that company and its
However, these kinds of data gathering and reporting
tools are not the only weapons in your arsenal. You can
also learn a lot about your site’s effectiveness through
user testing, focus groups and individual interviews. By
carrying out these exercises regularly it is easy to build up
a range of evidence of whether the changes you are
making to the site are working.
Of course, we shouldn’t just be measuring our
successes. We also need to track where things are going
wrong so they can be improved.
30 Measuring your successes and failures
Tracking your failures
All websites have weaknesses and it is important that we
know exactly what they are in order to do something
about them. What is more, websites develop new
weaknesses over time. As content is added, structure is
changed, and designs are tweaked, we introduce
problems into our sites that were not there before. We
need a system in place that continually monitors for these
failures so that they can be addressed quickly before
causing too much damage.
Implementing such a system does not need to be
onerous. It simply consists of three simple reviews that
should take place on a monthly basis. These are:
‣ Monthly user testing.
‣ Identifying dropout points.
‣ Analyse search queries.
Let’s begin with user testing.
Monthly user testing
Steve Krug’s latest book Rocket Surgery Made Easy takes
usability testing in a radical new direction that is both
perfect for identifying the constantly evolving weaknesses
in our sites and also inexpensive to implement.
Measuring your successes and failures 31
Steve Krug’s He proposes an ongoing programme of quick and
monthly user dirty user testing that takes place every month. The idea
testing is that on a set date each month you schedule a morning
programme of user testing with only 3 participants.
can be an Your entire web team watches the user testing and
excellent way analyses the results over lunch. By the end of lunch you
of identifying will have identified the biggest problems that need to be
problems. addressed before the next month’s testing.
This is a great approach for detecting and eliminating
problems on your website. It is:
‣ Lightweight – Only requiring one morning a month
for testing and debrief.
‣ Regular – Ensuring that it picks up on problems
that creep into the site over time.
‣ Action orientated – By the end of the session you
have a plan of how to tackle the failings in your
‣ Fixed – Because the user testing is always on the
same day every month, it is less likely to be
bumped for more ‘important‘ things.
‣ Prioritised – With only a short time to test and
debrief, the emphasis is placed on the most
important failings rather than being caught up in
Although user testing is useful, it should not be used
in isolation. Testing only 3 participants enables
lightweight and regular testing, but it does have its
limitations. Ultimately, 3 people are not going to be
representative of the whole and they are not interacting
naturally with the website. For that, you need to look at
32 Measuring your successes and failures
Identifying Drop Out Points
Website statistics can be very enlightening when used
correctly. Unfortunately, few people know what they are
looking for and so either give up try or never get past
page views and unique visitors.
One aspect of web stats that are particularly
interesting are exit points. That is, where do users leave
your site? This may help identify potential points of
weakness in the site and areas you wish to test in your
next user test session. Most statistic packages make it
easy to view a list of top exit pages. However, make sure
you are viewing pages with the top percentage of exits,
because popular pages will be exited more often. Look
for pages that are popular but are also exit points. These
are the biggest problem areas. Google Analytics actually
allows you to view pages that meet this exact criteria.
Google Analytics makes it easy to identify where users
are leaving your website.
Measuring your successes and failures 33
Once you have this list ask yourself the following
‣ Is this the final page in an action process? For
example, is this an order confirmation page at the
end of a purchase process? You would expect
pages such as this to be a typical exit point.
‣ Has the user visited any other pages before
exiting? If the user has viewed only this one page
then the chances are they were simply at the wrong
site. To make sure, check the dwell time. If they
were only on the page a few seconds before
leaving then it is fair to conclude they had arrived
by accident. Also, check how they arrived on the
site. If they came via a search engine, what search
term did they use? If the term wasn’t relevant to
your site then don’t worry.
‣ What type of content does the page contain? If, for
example, the page is a blog post on a standalone
topic, the chances are the user was more interested
in the content than anything else you offer. They
will read the one article and then leave. Obviously,
this is not what you want, and you should work
hard to encourage readers like this to venture
deeper into the site, however, the reality is that
pages of this type will have a higher exit rate.
‣ Is the bounce rate significantly higher than
elsewhere on the site? If so, this could indicate a
weakness in the page.
34 Measuring your successes and failures
If users have viewed multiple pages and then given up
on a particular page, it is a problem which needs
addressing. Possible problems could include:
‣ No obvious next steps – Is there a clear call to
action, ie, are you telling the user what to do next?
‣ Too many options – Too many choices can cause
users to give up.
‣ The content fails to convince – Is the content of the
page telling the user what they want to hear, or are
you saying something that alienates them?
‣ Too much content – Is the page packed with
large amounts of densely written copy? This can
put users off.
‣ The content is not relevant – They have arrived
from a search engine query and the page hasn’t
provided what they want. Instead of looking
elsewhere in the site they have returned to the
search engine to view another result.
Unfortunately, although they can be very good at
identifying problem pages, web stats are not so good at
diagnosing the cause. That is why it is important to user
test as well. Of course, another possibility is that a user
has given up simply because the site doesn’t appear to
have what they are looking for. That is where you need to
analyse search queries.
Analyse Search Queries
What users search for provides an excellent insight into
potential failures of your website. Take the top search
queries that lead people to a single page with an
exceptionally high bounce rate. Which of those queries
cause most of the bounces? What is the user expecting to
find on the page? The page must be relevant in someway
otherwise the search engine wouldn’t refer them.
However, perhaps the way you are presenting the
Measuring your successes and failures 35
information is wrong? Does the search term give any
indication of how you could be presenting things better?
What about the terms that create the highest bounce
across the whole site? Sometimes users find themselves
on a site that a search engine believes is relevant but
doesn’t directly address their issue. Could you re-tailor
your content to more directly address these search
However, it is not just external search engines you
need to be looking at. What about your internal search
engine? When a user arrives at your site after entering a
search term into Google, you can be sure that at least
some content on that subject exists. Otherwise, Google
would not have referred them. However, when a user
types something into your own site search, there is no
guarantee it will be a topic you have addressed at all.
This is a perfect way to identify content lacking from
your website. Repeat each of the top searches yourself
and look at the results. Are you addressing the search
terms directly? Are you addressing them at all? If not, you
have identified a weakness that needs addressing.
Rinse And Repeat
The secret to success when it comes to eradicating the
failures on your site is to establish a monthly cycle of
work. Each month you need to user test, delve into your
the failures web stats and analyse your search results. By combining
on your site these three techniques you should be able to establish a
is to programme of work for the coming month. By repeating
establish a this process, month on month, you can slowly evolve your
monthly cycle site so that any of its current shortcomings have less and
of work. less of an impact on conversion.
Business objectives are almost pointless without the
ability to track them. However, with a little creative
thinking it is possible to track almost everything.
However remember not to be seduced by the tool
itself and that tracking your failures are as important as
tracking your successes.
With this in mind, what actions should you be taking?
Action 1: Find the right tool
Once you have a list of business objectives the next
step is to work out what tool or technique will enable
you to track them. For some examples to get you
started make sure you read the suggested tracking
Action 2: Install Google Analytics
If you are going to track dropout points and search
queries you will need to have a web statistics package
running. Although there are a lot of solutions out there
I would recommend Google Analytics. It is free and
there is a lot of help available on how to use it.
Action 3: Arrange monthly user testing
I cannot stress enough the benefits of ongoing usability
testing. Arrange a day each month when you can test.
Also, read Steve Krug’s book Rocket Surgery Made
Easy. It will be invaluable.
Usability testing is invaluable for meeting business
objectives. That is why we are going to look in detail at
the subject in the next chapter.
Within certain circles of the web design community I am
described as a heartless commercial tyrant obsessed with
nothing more than the bottom line. I doubt that the first
two chapters of this book have done anything to dispel
However, although I think business objectives reign
supreme and that every website needs to generate return
on investment, that does not mean I am entirely
In the past I have been unfortunate enough to
encounter website owners that are so obsessed with
return on investment they put their business objectives
above the needs of their users.
Although in the short term this can generate returns, it
is a dangerous path to tread.
In this chapter I will demonstrate why it is so important
to focus on users needs. I will show how this benefits your
business objectives, generates return on investment and
what steps you can take to ensure users have a positive
experience on your site.
1. The cost of ignoring user needs
2. The profit of user testing
3. Bargain basement user testing
4. Connecting with users
5. Next Actions
Becoming user focused 39
The cost of ignoring users needs
It is very easy to become short sighted about business
objectives and fail to see the bigger picture.
Take, for example, an ecommerce website that wishes
to encourage more users to register when purchasing. A
short term view of this objective might be to force users
to register before they can place an order. However there
is a serious danger you will alienate users.
When users come to an ecommerce site they don’t
come to register, they come to buy a product. They
therefore see registration as a hinderance that benefits
the retailer more than themselves. This often leads to
users abandoning shopping carts and buying elsewhere.
Of course leaving to buy elsewhere is not the worst
that can happen. In a world of social media where every
users has a voice, the consequences can be much greater.
Users ability to damage your brand
You cannot Take the experience of computer manufacturer Dell. In an
ignore users attempt to meet their business objective of keeping
in a world prices low, they cut costs across their organisation.
where Unfortunately, one of the victims was customer service.
everyone has Eventually users became dissatisfied with their experience
the potential and this spilt over onto the web in a single post entitled
"Dell lies. Dell sucks.”
your brand. This single post galvanised a community of
disillusioned users and quickly got picked up by
mainstream media. Before long Dell had a serious PR
problem on their hands that haunts them to this day.
The lesson here is that you cannot afford to ignore
users needs when deciding how to fulfil business
objectives. If you do, ultimately it could cost you dearly.
40 Becoming user focused
Although disgruntled users can cause massive damage
to your website business objectives, happy users can
make them happen.
The power of happy users
If you can create a truly usable website that allows your
visitors to really engage with you, the returns can be
Among other things, happy users will:
‣ Drive more traffic to your site - If a user has had a
positive experience they will tell friends, tweet
about it and maybe even post to their blog. This
kind of customer testimonial is invaluable.
‣ Become site ambassadors - Happy users will not
just promote your site, they will defend it from the
criticism of others. Take for example Apple. Say
something negative about Apple and you will
attract angry fans faster than you can say
‣ Be more likely to help you - Happy users are more
likely to be amenable towards requests. This is
useful next time you have a user survey you want
‣ Be more forgiving when things go wrong - If a user
has previously had a positive experience of your
site, they are more likely to be forgiving when
something inevitably does go wrong. To create
So how do you ensure your users are happy? A part of happy users
the answer is to ensure your website is usable. To do this you must
you need to run usability test sessions as I mentioned in understand
Chapter 2 (Measuring your successes and failures). what they
However, usability testing does not just contribute to want and how
happier users. It also brings its own returns. they work.
Becoming user focused 41
Happy users can be
some of the most
of your brand.
42 Becoming user focused
The profit of user testing
Usability testing has traditionally been seen as an
expensive luxury that reduces profitability and extends
For years, usability testing took place in expensive
usability labs with one way mirrors, computer suites, and
video surveillance. Large numbers of test subjects were
required to provide statistically relevant data. Each
subject had to be hand picked to meet a certain
demographic profile. Testing was expensive and time
Usability testing does not need special equipment or a
This approach was effective but prevented most
companies from running sessions. Although a usability
consultant, testing in a lab, with demographically selected
subjects is nice, it is beyond the budgets and time frames
of most organisations.
However, user testing does not need to be that way. In
fact, it can be lightweight and inexpensive. It is also
something you can do yourself. It may not be as effective
but it is certainly a lot better than no testing at all. It will
still bring significant returns on your minimal investment.
Becoming user focused 43
The benefits provided by user testing cannot be
understated. Even the most lightweight of approaches
can have a profound effect on your web presence.
A usable site The benefits of user testing include:
will ‣ Increased chance of return visits - If a website is
encourage easy to use visitors are more likely to return. These
repeat users repeat users are the lifeblood of many websites.
who are They are more likely to respond to calls to action
considerably and become advocates of your site.
more likely to ‣ Reduced development costs - If done early in the
convert. development of your site usability testing can
identify problems that would cost a considerable
amount of time and money to fix later.
‣ Reduced running costs - If your website is intuitive
and easy to use, this will significantly reduce calls to
customer support and increase productivity among
staff who use the site.
With the benefits of testing so clear and the costs low,
the question becomes how do you run a session yourself?
Bargain basement user testing
In Chapter 2 (Measuring your success and failures) I
outlined Steve Krug’s low cost approach to usability
testing. I explained how he proposes a monthly
programme of user testing with only 3 participants. The
emphasis is on making usability testing easy and
lightweight in an attempt to ensure you actually do it.
Keep it lightweight and regular
The problem with traditional usability testing is that it is
too difficult to complete. Testing a greater number of
users is more effective as is ensuring they represent your
audience. However, this makes the process harder and so
the likelihood is you will not do it regularly.
44 Becoming user focused
Regular testing is important because it uncovers
problems previously hidden. Steve Krug explains it like
The first three users are very likely to encounter nearly
all of the most significant problems... Also, since you
will have fixed the problems you uncovered in the first
round, in the next round it’s likely all three users will
uncover a new set of problems since they won’t be
getting stuck on the first set of problems.
You maybe concerned that testing only three users will
not uncover enough problems to make the testing
worthwhile. However, this proves not to be the case.
According to Jakob Nielsen’s research 3 users will
uncover almost 75% of the usability problems on a site.
Beyond that the additional return on investment is
The more users you test the less return you get from
each additional users.
Becoming user focused 45
Combined with the fact that 3 users allows you to test
and debrief in a morning, it seems that testing more often
doesn’t generate enough returns to justify the
Of course, the time spent on usability testing is not
just the session itself. Some preparation is required too.
Preparing for your test session
Preparation for a user test session falls into four areas:
‣ Finding test subjects
‣ Setting up the test environment
‣ Encouraging stakeholder attendance
‣ Preparing the script
Let’s look at each in turn.
Finding test subjects
Although it When you think about usability testing it is natural to
is great to presume you should test the exact target audience.
test your However, in reality it does not much matter who you test.
target Although there are exceptions (such as a website aimed
audience it at a particularly elderly or specialist audience) most
people will encounter the same usability hurdles no
matter what their background.
be at the
expense of Although it is great to test your target audience it
the amount should not be at the expense of the amount you test.
you test. With recruitment now reduced to grabbing any
random individual not involved with your company or the
website, you can turn your attention to where to test.
Setting up the test environment
Setting up your testing environment is amazingly simple.
As I have already said you do not need a usability lab, one
way mirrors or even a video camera (although this helps).
All you need is a room, a computer connected to the
46 Becoming user focused
internet and two chairs (one for you and one for your test
Ideally the test environment should be somewhere
quiet where you will not be disturbed. An office or
conference room are ideal.
Because you are facilitating the session it is going to
be impossible for you to remember everything that is
said. That is why it is also important that your test
environment has some means to record the session.
My preferred approach is to use screen capture
software that also records audio and possibly even video
via your computers webcam. For windows users I
recommend Camtasia. This general purpose screen
capture tool is also available for the mac. However,
instead I would recommend mac users use Silverback.
This tool is made specifically for recording usability test
sessions and is considerably cheaper than Camtasia.
Use a tool like Silverback to record your usability testing.
Becoming user focused 47
However, you don’t just need to be able to record the
sessions. You also need your stakeholders to be able to
view them live.
Encouraging stakeholder attendance
Nothing helps understand the importance of usability
more than watching a user struggle to complete a task on
your website. Until you see it first hand it is impossible to
appreciate how poorly a badly designed site can reflect
on your brand and profitability.
That is why it is so important to ensure as many
internal stakeholders can watch the usability test sessions
as they happen. This not only includes the website
production team but also senior management and other
The best way I have found for this kind of remote
viewing is GoToMeeting. Although primarily a screen
sharing tool it is ideal for hearing and seeing what a user
Gotomeeting allows stakeholders to watch the user
sessions live from another room.
48 Becoming user focused
Once the time and place has been set, and users
found, all that remains is to identify what to test.
What are you going to test?
Before you walk into the first user session it is important
to have a clear vision of what you want to test. These
‘tests’ need to be turned into tasks that users will be
asked to do. However because user testing is an intense
experience for participants a session should not last more
than 30-40 minutes. In my experience this equates to 3 or
Establishing what these tasks are will vary. However,
personally I draw tasks from three main sources:
‣ A point of contention - Sometimes there is
disagreement between stakeholders over the best
way to approach a particular element of the
website. For example, the marketing department
may make a suggestion that the web designer
disagrees with. In such situations usability testing
can be an effective way of identifying the best way
forward. Testing with real users gives them the
‣ Following up on usability fixes - When a usability Once you fix
problem identified in a previous session has been a usability
fixed it is good to test again in order to ensure it problem
has really been fixed and that it hasn’t created always test
additional problems. again to
‣ An investigation of potential problems - As I ensure the
suggested in chapter 2 (measuring for successes
and failure) web statistics can be a good way of
identifying potential problems. However, you will
also receive customer feedback about potential
usability issues. Before these are fixed they first
need to be confirmed as serious issues and that the
problem is fully understood. Usability testing is the
best way to achieve this.
Becoming user focused 49
Once you have identified your problems it is relatively
easy to work out a task. For example, if users are
dropping out on a particular page ask participants to
complete a task which involves navigating via that page.
Equally, if users are adding items to a shopping basket
but not purchasing, ask participants to place an order and
see what happens.
there was disagreement about
a new feature for the site...
I really don't think
users are going to find
that feature useful
OK. Why don't
we test it and
User testing can be a great way to diffuse arguments
Once you have a list of your tasks it is simply a matter
of carrying out the actual testing.
How to run the test session
The secret to running a successful test session is the
facilitator. This is the person who interacts directly with
the participant. Unless you are using an expert, this role
will probably fall to you.
50 Becoming user focused
The facilitators role
Whoever your facilitator is they need to understand their
role. The position has three responsibilities.
‣ To intervene when necessary - There is a perception
that the facilitator should never intervene to help a
struggling tester. However, this is counter
productive. If a user becomes completely stuck,
show them how to continue. This provides an
opportunity to discuss the problem and allows a
chance to discover other issues deeper in the site.
‣ To lead the tester through the usability script - It
always pays to prepare a usability script
beforehand. This outlining what you intend to
cover. It is the role of the facilitator to guide the
tester through this script.
‣ To encourage the tester to communicate - Many
users sit in silence struggling with a site unless
encouraged to speak. The facilitator should
challenge users to think aloud. Ask open ended
questions and ask what they are thinking.
The test participant needed
Why did you click
there? what was
The facilitator needs to encourage the participant to
explain what they are doing.
Becoming user focused 51
However, the facilitator also has one more crucial role,
to put the user at their ease.
Putting participants at their ease
If a user User testing can be an intimidating experience for the
thinks you participants. This can prejudice results because
were involved participants are nervous, self conscious or have a desire
in the to please. Fortunately there are things you can do to put
website they people at their ease.
will be more ‣ Introduce the environment and people - Introduce
reluctant to yourself and anybody else in the room. Explain the
criticise it. presence of any cameras and tell them who will see
any recordings you are making.
‣ Explain that you are testing the website not the
participant - Many participants feel they are the
one’s being tested. Explain that there are no wrong
or right answers and that the idea is to make the
website easier for them to use.
‣ Distance yourself from the project - Participants
don’t like to offend and so sometimes tell you what
they think you want to hear. The only solution to
this problem is to lie! Tell them you are not involved
in the project so they can be as rude about the site
as they like.
‣ Start with some simple questions - Instead of
jumping straight into your tasks ask the participant
a little about themselves. This provides you with
useful background information and helps to relax
participants because talking about yourself is easy.
‣ Offer participants a drink - Its amazing how offer
participants a coffee can change the dynamic of a
session. It goes from being a formal environment to
something much more relaxed and social.
‣ Test in people’s homes - Instead of testing at your
office consider going to their home. They will feel
more comfortable and you will have an interesting
insight into how they normally access the web.
52 Becoming user focused
Of course, participants are not the only people to get
nervous. You may also feel nervous before running your
first session. However, you will quickly discover that
usability testing is remarkably easy. The hard part is what
What to do with the results
Once the usability testing is over it is time for the entire
team to gather together over lunch and decide on what
problems to fix over the next month.
The problem is that user testing can feel
overwhelming, especially the first time you do it. Many
people go into usability testing hoping that it will validate
the approach they have taken. Instead it often does the
However, although things can often appear to be a
disaster they are rarely as bad as they seem. Not every
criticism requires a response and not every problem
needs fixing. The key is to filter and prioritise.
Filter out background noise
The first step is to weed out the distractions and the Don’t allow
first type of distraction is design related. In usability users
testing participants often comment on aesthetics. This is a suggestions
distraction for two reasons. First, the aim of usability for new
testing is not primary to address visuals. Second, design is functionality
very subjective and comments like “I do not like the to derail the
green” have little value. Unless the vast majority of testers
are making the same comment, you can ignore anything
said about design.
The second common distraction is suggestions for new
content. Comments such as “wouldn’t it be good if the
site did this” do not help evaluate what is currently there.
The scope of the project was defined much earlier in the
process and if new content or functionality is added at
this stage it will lead to scope creep.
Becoming user focused 53
Design is very subjective and so don’t be too concerned if
participants react negatively to the site’s aesthetics.
I am not suggesting that ideas about content are
worthless. As with design suggestions, if the vast majority
are suggesting something you have to take their
comments seriously. Ask whether you have missed
something fundamental or whether it is simply a nice
The final filter is that of return on investment.
Sometimes a problem cannot be fixed without investing a
disproportionately high amount of money compared to
the return received. In many cases this is because of
existing technology or business processes which cannot
54 Becoming user focused
Although there are things we can do to minimise this
inconvenience we are ultimately limited by the realities of
Once the list of issues has been reduced, the final step
is to prioritise.
Setting your priorities
With only a month until your next test session there is You don’t
only so much that can be done. It is therefore important need to solve
to prioritise. all your
This prioritisation is based on two factors; the usability
complexity of the fix and the seriousness of the problem. issues in one
I would encourage you to search for quick wins and deal go. Prioritise
breakers and prioritise these. and address
‣ Quick wins are minor issues that are so easy to fix, them over
it is silly not to. A good example of this might be a time.
poorly named link that made a user pause for
thought. It didn’t stop the user completing the task
and so cannot be considered crucial. However, it is
so simple to rename the link that there is no reason
‣ Deal breakers are the issues that prevents users
from proceeding further. A classic example is a user
who cannot find the checkout button on an
ecommerce site. Without it they are unable to
continue and are forced to give up. They are
priority issues that simply have to be fixed.
By prioritising these two areas you will have to ignore
other smaller issues. However, these can be addressed in
future sessions. You don’t need to solve everything in one
go. Rome wasn’t built in a day.
By following the advice given here and in books like
Rocket Surgery Made Easy you will make your site a much
less abrasive experience for users and increase their
satisfaction. However, avoiding upsetting your users is a
fairly low bar.
Becoming user focused 55
Connecting with users
Usability is like cooking edible food. Its a minimum
requirement rather than the aim. Nobody goes to a
restaurant because they serve edible food. They go
because the food is delightful, the ambience is perfect
and the staff are welcoming.
Food should be more than merely edible. It should taste
delicious. Equally your website should be more than just
usable. It should delight.
If you want to create happy users (with all the benefits
that provides) you need to both engage and delight
them. This begins with establishing the right relationship.
The right relationship
The web is an impersonal place. It lacks many of the
facets that makes human communication so rich. There is
often no body language, tone of voice or facial
56 Becoming user focused
Although video and audio content is becoming more
prevalent, the majority of website owners communicate
with their users primarily through text.
This can make websites feel cold and dispassionate.
Website owners appear not to care about their users.The
problem is further amplified with corporate speak and
With such a backdrop it is hard to turn visitors to your We do not
site into passionate, happy and engaged users. As a building
result few website owners really derive the full benefits of relationships
having committed users. with
What many are missing is that people like to engage websites or
with other people. We don’t build relationships with corporations.
websites or corporations. Neither do we derive happiness We build
from them. relationships
Its easy to become frustrated with a website or with people.
company. It’s much harder to become passionate about
If we are passionate about connecting with our users
in order to truly excite them about our products and
services we need to engage them as one human to
another and not as a faceless corporation or website.
This is achieved in a number of ways.
‣ With a personal and appropriate tone of voice
‣ By being human and real
‣ Through dialogue rather than monologue
‣ By building community
This begins by establishing the right tone of voice.
The right tone of voice
If your website was a person, who would it be? Maybe it
is a business titan like Alan Sugar or a celebrity like
Becoming user focused 57
The point is that your website should have a character.
Preferably one that is reflective of your culture and able
to build a relationship with your users.
Your site’s character dictates how you write, what
images you chose and how you interact with users.
It will dictate everything from the customer emails you
send to the labels on buttons. Everything should have a
touch of your particular character.
The emails of print company moo.com are full of
character in the form of ‘little moo’, a fictional robot
that deals with your order.
58 Becoming user focused
However, establishing your character is not as easy as
it would first appear. On one hand your character has to
be representative of your organisations culture. If it is not
then the character feels false and is difficult to maintain.
On the other hand it needs to be a character that can
effectively communicate with your audience. A character
your audience will like and respect.
Unfortunately, many organisations have a culture that
is not particularly compatible with their audience.
However, if you want to build a community of passionate
users your culture will need to adapt.
That doesn’t mean your site’s character has to be the
same as your users. It just means your character has to be
one that your users like and respect.
The innocent smoothies company do a great job at
portraying a character that is fun, approachable and
socially responsible. They manage to engage the young
festival goer who attends the music events they sponsor
and a more elderly audience who knit woolly hats for their
drinks to raise money for charity.
Innocent smoothies engage a wide audience through its
socially responsible and yet fun character.
Becoming user focused 59
Having a single character doesn’t mean you cannot
appeal to multiple audiences. It just means that your
website seems more human.
Learning to be a human
One of the biggest elements of being human is to be
genuine. One of the best ways to demonstrate that online
is by admitting when you make mistakes.
I am constantly frustrated by how many companies put
up this infallible facade. Its like the entire place is run by
machines that never make mistakes.
They are not fooling anyone. We all know that
companies are not run by perfect robots but by fallible
humans. We can except failings but we cannot tolerate
people refusing to take responsibility.
What is particularly interesting to me is that the
occasions when you mess up can also be the best
opportunities for turning the most harden cynic into a fan.
Photo sharing site avoided a major community backlash my
apologising for mistakes that it made.
More than once I have heard users tell tales on twitter,
facebook or their blog of situations where a company
online had screwed up but then bent over backwards to
make things right. These users have been so impressed
by the customer service they have received it has led
them to post about it and telling others how great the
60 Becoming user focused
Sometimes it doesn’t even need to cost the company
anything. A sorry is often sufficient. In the early days of
Flickr they were plagued with outages which left paying
customers unable to upload or access their photography.
As you can imagine, they were far from happy.
However, the potential backlash was largely defused
by a single blog post entitled “Sometimes we suck” in
which Flickr admittedly responsibility, apologised and
explained what they were doing to rectify the situation.
For many users this was the point when they became
dedicated fans of the site.
Of course, showing the humanity of your organisation
online is a lot more than saying sorry. It’s about showing
the full breath of human emotion from a sense of fun to
passion and enthusiasm.
Often the easiest way of showing a human side is to
allow people within your organisation to speak with their
own voice directly to the customer. It is easy to get angry
with Microsoft. However, it is much harder to criticise a
programmer when you understand the context of his job.
Microsoft allow you to see behind the corporate facade
and meet its employees on its blog Channel 9.
Becoming user focused 61
Another crucial part of being more human is to start
having conversations with your users rather than just
broadcasting at them like a TV or radio channel.
Speak with users, not at them
Traditionally marketing departments have dealt with
broadcast mediums such as billboard advertising or TV
commercials. However, the web changes all of that. It
allows a two way dialogue and that is something many
marketeers have struggled to grasp and adopt.
Since the launch of the world wide web the vast
majority of websites have focused on broadcasting
information to consumers. However, more recently we
have seen a shift that have allowed users to talk back
through tools such as Twitter and Facebook.
The CEO of sales retailer Zappos uses twitter as a way
to really engage with his customers.
This is a good thing because people like to have their
voices heard. As a website owner if you take on board
users comments and talking directly to them, you will see
incredible results. If you listen and act based on user
feedback those users will begin to feel like the website is
62 Becoming user focused
as much theirs as it is yours. Once they have made that
mental leap they will be as keen as you to see the site
succeed. This is the point where they move from being
passive consumers to active evangelists.
Constantly look for ways to give your users a voice.
Whether it is through commenting on your corporate
blog, a forum or by allowing them to submit user
generated content, you need to make users feel as
engaged as possible.
Although this kind of two way communication is good,
an even better option is to move to a community based
Building a community
Building online communities has enormous benefits to If done right
your business if done right. Communities bring with them having a
a sense of belonging that most people deeply desire on a community
sub conscious level. It creates a loyalty that will keep associated
users coming back again and again. with your
Communities can also be mobilised to have a website will
significant impact and act as the most effective marketing transform
tool you will ever have. your
Communities also work as a wonderful support business.
mechanism for new users, significantly reducing the
customer support burden for you.
Building a community is not particularly easy and not
something I can cover here. It takes time, commitment
and a genuine interest in community members. That said
if done right, it will transform your business.
Building community, engaging users and establishing
the right relationship will have a profound impact on the
happiness and enthusiasm of your users. However, there
is one last detail that will complete the picture - delightful
Becoming user focused 63
In the early days of the web people used to ‘surf’ for
‘fun’. This rarely happens anymore. With the exception of
Youtube and a few others, most websites exist to serve
up information to users or allow them to complete tasks.
The web has become a utility and not a pleasure.
This provides a unique opportunity to stand out from
the crowd. If you are looking for a way to make your site
more memorable and encourage users back then make
your users smile and try to delight them.
Its possible to add some fun to any website no matter
how dull and uninspiring the subject matter. Take the
HTML email delivery service Mailchimp.
Sending out bulk email newsletters is not the most
exciting of activities. They are boring to create and
involve the user completing a number of relatively
elements of fun to
their website that
makes a dull job
64 Becoming user focused
Mailchimp have set themselves apart from the
competition by adding humour to the interface including
funny one liners delivered by the ‘chimp’ and small design
details many might not even notice.
You may wonder what the point is if hardly anybody
notices. Where is the return on investment in this? The
return is two fold. First, that moment of delight for those
who do notice sets you immediately apart from the
competition by humanising your site and ensuring it is
memorable. Second, they tell their friends about this
hidden easter egg who in turn tell their friends. I have
heard Mailchimp mentioned time and time again because
of these added delights, just as I am now mentioning
them to you. When next you need to send out an email
newsletter, who are you going to think of?
Of course, adding humour like this is not appropriate
to every site. However it is always possible to add
something. Some hidden gem that makes somebody
smile or have a ‘wow’ moment. These are the details that
set aside the truly great sites.
Best of all these kinds of details do not require huge
amounts of work. Many are a matter of minutes to
implement. The trick is not implementing them, but
realising their importance in the first place.
The mail chimp’s little comments are enough to make the
most serious of us smile.
This chapter is the longest in the book for good reason.
Putting users first offers the greatest return for your
business over the long term but doing so consistently can
be challenging. I recommend you begin this journey by
taking the following steps:
Action 1: Arrange monthly user testing
Following Steve Krug’s advice about running monthly
user testing means that users are constantly at the
heart of your thinking. By testing with real users every
month and looking for ways to improve your
experience you can never forget to put them first.
Action 2: Start truly engaging your users
Now is the time to break away from talking at your
users so that you can begin engaging with them. Put
opportunities in place where users can feedback to you
whether through comments, twitter or facebook. The
more you listen to users the more engaged they will be.
Action 3: Look for ways to delight
Finally seek out opportunities to differentiate yourself
from the competition and encourage users to mention
you to friends. Every website can have ‘delighters’ that
grab users attention and just make them smile.
The positive results of delighting your users cannot be
overstated. However, it is important to remember that it
is not the end goal. We do so to encourage more users to
complete your calls to action, so increasing your ROI. The
problem is that if your calls to action are not right it won’t
matter how happy your users are. It is getting these calls
to action right that is the topic of the next chapter.
calls to action
I have talked about Apple several times already in this
book. It seems to be a common theme among web
designers. We love to praise Apple. However, my first
experience of an Apple store was not a pleasant one.
After months of listening to Apple fans tell me how
great Apple products were I decided to buy my first
Macbook. All I had to do was walk into the store and buy
it. Unfortunately, this was harder than you would expect.
The problem was that the store didn’t have the normal
visual queues I expected to find. There were no boxed
products to pick up and take to the cashier. Worse still,
there was no cashier! Where was I meant to pay?
After standing there looking like a fool for a few
minutes I wandered over to the genius bar. No, that
didn’t look right. People were getting tech support, not
purchasing. What was I meant to do?
Eventually I asked an assistant and he dealt with my
purchase. However, I felt foolish asking and left the store
What I needed was some indication of what I was
meant to do. I needed a call to action.
1. Why calls to action matter
2. Deciding on your calls to action
3. Motivating users to take action
4. Designing your calls to action
5. Following through on actions
6. Next Actions
Creating your calls to action 69
Why calls to action matter
A call to action is the point at which you ask users to
respond in someway. That could be something as major
as placing an order or as trivial as clicking a link to
proceed deeper into the site.
Everybody presumes that calls to action are primarily a
business tool. This is true. However, as my story about the
Apple store demonstrates they are not only for the site
owner’s benefit. They also help the user too.
Where do you place your order in an apple store? Where
is the call to action?
70 Creating your calls to actionn
Why users need calls to action
You might think me foolish for being so confused by an
Apple store. However, I can guarantee that at some point
you have been to a website where you felt unsure what to
do next. This is often because the site owner has failed to
give you any guidance as to your options to progress.
They have failed to give you a call to action.
The web is meant to be a massive interconnected
network of pages. However, in reality a large number of
those pages are dead ends. This gives users only two
option, leave or go backwards. As humans we hate going
back over old ground. Therefore, the consequence of no
calls to action is obvious. Users leave feeling unsatisfied.
Every link on wikipedia is a micro call to action drawing
the user deeper into the site.
A good call to action makes your site sticky. It is draws
the user onwards towards a final goal. However, it doesn’t
always need to be flashy or crude. In my opinion, one of
the sites with the most effective call to action is
Wikipedia. Every link is a call to action that draws you
ever deeper into the site. Once you start reading
Wikipedia entries it is hard to stop.
Calls to action are not just about avoiding dead ends
that leave users feeling unsatisfied. It is also about
helping users complete a key task. Take the browser
manufacturer Mozilla. If users visit the website
getfirefox.com you can pretty much guarantee they want
to download a copy of Firefox.
Therefore, it is hardly surprising that the website
sports possibly the largest call to action known to man!
Mozilla know that the majority of their users only want
to do one thing, download Firefox. They make that as easy
as possible with an obvious call to action.
However to pretend that calls to action are only for
the benefit of users would be simplistic. They are also
beneficial to website owners.
The business beneﬁts of calls to action
Calls to action offer many benefits to business too. They
encourage users to become active participants with your
brand, product or site. They also allow you to track
whether the investment in your site is paying off.
Moving from passive to active
As any sales person will tell you, in order to bring a Encourage
potential customer to the point of sale you need to move users to take
them from a passive state (a passive interest in your a small action
product or service) to an active state in which they will before asking
take some action to learn more about what you have to them to make
offer. In other words it is not enough for them to agree to a bigger
listen to your sales pitch. They have to actively want to commitment.
A call to action attempts to move the user from this
passive state to a more active role. Once a user has taken
some kind of active step (even if it is only to sign up for a
newsletter) they have mentally transitioned to taking an
active interest in your offering.
In short, calls to action are an integral part of the sales
process. However, they are also an important aspect in
Tracking return on investment
In chapter 2 (measuring your successes and failures) I
talked about success criteria. The idea of having
measurable ways by which you judge the success (or
otherwise) of your website.
It is these success criteria that validate the decision to
invest in your site and provide justification for future
Your calls to action are the steps that the user must
complete in order to meet the success criteria. Without
the user taking these steps you would have no way to
track the effectiveness of your site.
Calls to action can be anything from completing an
enquiries form to printing a coupon. However, whatever
they are, it must be something that can be trackable so
success can be measured.
The restaurant chain Prezzo has a money off voucher as
their call to action.
Although we now know that calls to action should be
trackable, we haven’t discussed how to pick your call to
action. This is obviously a key part of getting users to take
Deciding on your calls to action
In this chapter I have given a couple of examples of calls
to action. However, those were just the tip of the iceberg.
A call to action is anything that asks the user to move
from being a passive consumer to an active participant.
This could include something as everyday as the
common hypertext link. However, it could also include:
‣ Purchase buttons
‣ Signup forms
‣ Download links
‣ Checkout processes
‣ RSS subscription
‣ Telephone numbers
‣ Email links
‣ Contact us forms.
The list could go on.
Notice that not all of those lead to a sale. There is
often a perception that calls to action only apply to
ecommerce sites. This is not true. Calls to action can be
transactional. However, they could just as easily be a user
requesting information or signing up to a web app.
The calls to action on Boagworld are to encourage users
to subscribe to the podcast.
Also remember what I said earlier, every page should
have a call to action. Users should never be left at a dead
end not knowing what to do next. That said, not all calls
to action are equal.
Page level calls to action
In Chapter 1 (Use business objectives to inform decisions)
I wrote about micro and macro business objectives. The
same approach can be used for calls to action. When
writing a list of your site’s calls to action you need to think
about those for individual pages as well as site wide.
These ‘micro calls to action’ aim to do two things.
One, to keep the user on the site and, two, to move them
one step closer to your site-wide calls to action.
Boagworld also have page specific calls to action designed
to keep users engaged with the site.
One method to ensure every page has a call to action
is to use a content template, known as ‘page tables’ in
content strategy. A page table asks the writer of a page
to answer a series of questions. These include:
‣ Who is the page aimed at?
‣ What is the main message the page should
‣ What will the user learn on this page?
‣ What next step should the user take?
This last question will ensure the writer always thinks
about what call to action a page should have. This
ensures the user is always moving one step closer to your
Although these page level calls to action exist to move
the user toward a site wide action, that does not mean
you can only have one site wide call to action.
Prioritising your site-wide calls to action
It is not unusual for a website owner to have several site Avoid
wide calls to action such as buy a product and signup for overwhelming
a newsletter. In such cases it is important to prioritise. users with
Begin by establishing your primary call to action. What conflicting
must users do above all else? Knowing this allows you to calls to
design the site in such a way that the user is in no doubt action.
about what they should ultimately do.
Of cours,e on an extremely large site with a diverse
audience it might be necessary to have multiple primary
calls to action across different sections and sub sites. This
is fine as long as they are not competing for a user’s
attention. If the user is faced with conflicting choices they
often choose not to choose.
Your secondary calls to action should be less
prominent so as not to distract from the main goal.
However they do play an important role. They exist to
move the user towards the ultimate objective. For
example asking people to signup for a newsletter is a
much smaller step than purchasing from the site. Once a
user has signed up for a newsletter they are more likely to
make a purchase later. Just don’t allow the newsletter
signup to overshadow purchasing.
Apollo shows the primary call to action (signup) through
the use of colour. However its secondary call to action
(take a tour) helps move users towards signing up.
This problem of overshadowing is more common than
one would think. Sometimes we add these secondary
calls to action onto our site without even thinking about it
and inadvertently distract the user from what you want
them to do. For example, have you added a retweet icon
or facebook button to your site? These are calls to action.
However, what would you prefer users to do, tweet or
buy your product? Don’t put actions on your site that will
draw the user’s attention away.
Once you have decided on your calls to action the
next question is how do you make them enticing enough
to encourage users to click?
Motivating users to take action
There has been a lot written about how to make You cannot
somebody complete a call to action. The truth is that you force users
cannot make somebody do something they don’t want to to complete a
do. However, it is possible to nudge users in the right call to
Personally I use 5 techniques that encourage users to However you
complete a call to action. These are: can nudge
‣ Carefully wording calls to action them in the
‣ Focusing on feeling and benefits, not features right
‣ Remove any sense of risk the user might feel direction.
‣ Offer a carrot but also wield a stick
‣ Use the power of peer pressure
Probably the most well known is the need to carefully
word your calls to action.
Carefully wording calls to action
How calls to action are worded makes a huge difference
in how users respond. It is an area that has been
extensively researched and written about. There are also
many people far more qualified to write about this highly
skilled area than me. That said, I can point you in the right
direction with four practical suggestions:
‣ Use active language - Begin each call to action with
an active verb: learn, place, add, submit, get,
modify, edit, etc. These are far more likely to
motivate users into action
‣ Make it free - Research has shown that there is a
real power in the idea of free. Offering something
for free removes risk and we all want something for
nothing. For example: “Get a free market analysis
of your home.”
‣ Ensure it is tangible - Avoid using abstract concepts
or meaningless figures. Use something tangible
that users can relate to in their own lives. For
example instead of writing “Save a child's life for
only £2” write “Save a child's life for less than a
‣ Personalise it - By using the words “you” or “your”
users are more likely to apply your call to action to
themselves. For example: “Learn how you can
afford a bigger house than you thought!”
37 Signals use both active language and the power of free
to encourage users to take action.
Creating association with the call to action by
personalising it is very effective. However, you can go a
step further by encouraging the user to consider how
responding will make them feel.
Focus on feelings & beneﬁts, not features
We often tend to focus on the features our products or
services offer. For example we talk in terms of numbers,
specifications and services. However, an effective call to
action should focus on feelings and benefits, not just
Describing your product, service or call to action is not
enough. You have to make users want it. To do that you
need to describe the benefits it will have to them
personally. It is not enough to say how it will benefit their
customers or business. You have to talk in terms of
benefits to them individually.
For example, instead of having a call to action that
reads “our product will increase customer satisfaction by
80%:” write “Our product will reduce the number of
customer complaints you have to deal with.” Do you see
the difference? The first focuses on the benefits to
customers the second on the benefit to you.
However, you can go even further and engage with
people’s emotions. Instead of just talking about benefits,
talk about feelings. For example you could reword the
example above to read “Do you hate dealing with
customer complaints? Our product will increase
satisfaction so reducing the stress of dealing with angry
Earlier I talked about the power of free. I suggested
that part of the reason free was so appealing was because
it reduced the risk associated with responding. However,
free is not the only way to reduce risk.
This website focuses
heavily on how the
user is feeling and
how they can make
Remove the risk
One of the major reasons users are resistant to
completing a call of action is because they associate a risk
with it. These perceived risks will vary according to the
call to action. However some examples might include:
‣ “If I buy this product and don’t like it I might not be
able to return it.”
‣ “If I sign up to this newsletter they may send me
junk mail or sell my email address.”
‣ “If I want to leave this web application I might lose
‣ “What if this is a scam to get my credit card
‣ “I am worried about the privacy of this application”
If you want users to complete these calls to action you
must reassure the user.
Sites like Zappos do a superb job at this with a 365
day return policy and a no quibble money back
Zappos removes risk with an amazing return policy
You may think that Zappos extreme promises are
excessive. However, the web is an impersonal
environment and users have little information to go on
when judging your trustworthiness. This makes them
especially hesitant and so it is necessary to go the extra
mile to remove risk.
Associate with your calls to action promises addressing
users concerns. Whether it is outlining your return policy
or giving a clear statement about your approach to
privacy, make sure the user has no reason to worry.
Sometimes it isn't possible to completely eliminate all
of the users concerns. In such cases the carrot and stick
approach can help.
Offer a carrot but wield a stick
Website owners have successfully been using the carrot
and stick approach to encourage users to complete calls
to action for sometime. The carrot is incentives and the
stick is limitations.
The carrot: incentives
If you want If you have ever purchased a car you will know the power
users to of incentives. You don’t need to show much hesitancy
complete a before the car dealer will start throwing in sat nav
call to action systems, alloy wheels and all kinds of ‘extras’ to make a
throw a sale. Although compared to the price of the car these
incentives are inexpensive, they are often enough to push
us over the edge and make us feel we are getting a good
The same principle works on websites. If you want
users to complete a call to action throw a bonus for free.
Maybe that is a voucher off their next purchase, a free
ebook or a chance to win a prize.
My favourite example of this was the campaign to
elect Barak Obama. If you made a donation over a certain
value they sent you a free t-shirt. What was so genius
about this was that this encouraged people to increase
their donation to get a t-shirt that cost hardly anything
while persuading people to become a walking billboard
for the Obama campaign.
By offering an incentive the Obama campaign was able to
increase the level of donations.
Think about your calls to action and ask yourself
whether you could offer some form of incentive to
motivate people. Failing that, there is always the stick.
The stick: Limitations
The stick comes in many different forms however it is
normally some form of limitation. This might be a
limitation in availability (“only 30 seats available”) or a
limitation in time (“only one day left to book”). Of course
you could combine the two (“the 30 tickets available at
early bird price which must end next Tuesday”).
We see this stick being used all the time and for some
reason we always fall for it. The more limited the
availability, the more we want it. The more limited the
time, the less we want to miss the opportunity that
currently presents itself.
One of the best examples of this was when the
Nintendo Wii was launched. Because Nintendo
announced there was limited stock it led to a frenzy of
customers queue for hours to get their hands on the
precious console. This is also a trick Apple has used time
and again. Its strange, because despite knowing this trick
I will still stand in line to buy whatever Steve Jobs tells me
Events organisers Carsonified create a sense of urgency
for their workshops by limiting availability and creating a
sense of urgency.
These techniques play offhuman psychology. However,
sociology plays a role too.
Use the power of peer pressure
We are highly influenced by the behaviour of others and
this can be a useful tool in encouraging people to
complete calls to action.
We all know that testimonials are an effective tool in
persuading users to take action. Admittedly, these are
becoming less effective as users become more cynical
and doubt their independence. However, website owners
are responding by posting video testimonials,
unmoderated reviews and tweets about their products to
add weight to the reliability of these testimonials.
However, testimonials are not the only option available
and this social pressure is not just limited to sites selling
Have you ever been walking down the street and seen
a crowd of people gathered around ‘something’? It is very
hard to resist the temptation to see what they are looking
at. The same is true online. When we see a lot of people
doing something, we are drawn towards the same thing.
Where testimonials focus on great feedback from a small
group of select individuals, this approach uses the power
of the crowd.
One example of this was a website that campaigned
against Microsoft’s changes to Outlook. Their call to
action was to ask people to tweet their support. To
reinforce the popularity of the campaign and pursued
others of its value, they posted the profile pictures of all
those who tweeted as the background of the site. This
mass of people left you with one overwhelming feeling
“this is important, I should join in.”
This site demonstrated the level of support it was
receiving by posting people’s profile pictures as a
Let’s say one of your objectives was to get more
people involved in your forum. One approach would be
to post lots of conversation starters from lots of different
accounts. This gives the impression of lots of activity and
that in turn attracts more people. I have even known
some sites to pay people to participate in the early days
of a forum to attract a crowd.
This social pressure phenomena works in other ways
too. Have you ever seen a website that has said
something like “people who looked at this also looked at
this?” This plays off our ‘herd mentality.’ If I am looking at
this page or product and so did these other people then I
will probably like what they looked at next.
Amazon uses our natural desire to fit in with the crowd
to sell us products.
I have of course made one large assumption so far. I
have assumed that users have seen the call to action in
the first place. That is where good design comes in.
Designing your call to action
It is not my aim to teach you how to design. I am
presuming that you are a website owner and not a
designer. The chances are you don’t have the time or
inclination to learn design principles.
I do however want to share with you some basic
techniques used in call to action design. This is so you
understand some of the design decisions your designer
will be making. Some of what I share is fairly obvious.
However, other parts may appear counter-intuitive.
Lets begin with position.
Position is everything
When it comes to calls to action position is everything.
Your designer will want the call to action to have a
prominent position on the page. This will mean placing it
high on the page, but also at the bottom. Users often
look for the next step once they have scanned the page
so putting a call to action at the bottom is essential.
Picsengine focuses users on their call to action by
placing it centrally on the page.
On the surface this all sounds fine. However, it can be
painful handing over key real estate to a call to action
when there is so much more you want to say.
That brings me nicely onto removing clutter.
Your designer will encourage you to strip as much content
away as possible to focus the user on your call to action.
Many website owners resist this kind of extreme measure
but it is essential. Compare the Google and Yahoo
homepages. One has so much clutter on it that it is hard
to spot search. The other focuses the user immediately on
the search box.
Google focus the user on their primary call to action (to
search) by removing anything that might distract.
Ultimately, users only have limited attention. Make
sure they are giving that attention to the right things.
A related principle to removing clutter is white space.
Designers bang on about white space all the time. They
are referring to the empty space on the page. They will
almost certainly want to surround your call to action with
as much white space as possible. This is because the
more empty space around it, the more your eye is drawn
to the call to action itself.
PlanHQ draws attention to their calls to action by
surrounding them with empty space.
Many website owners have the overwhelming desire
to fill empty space. However, by doing so you are
effectively camouflaging your call to action. Instead you
want it to stand out like a groom on his stag night who
has been made to go paintballing in a pink jump suit.
(Trust me, they really stand out).
Talking of pink jumpsuits in a forest, I should also
mention colour contrast.
Colour is another useful tool in drawing attention to a call
to action. This is where some flexibility maybe required
on your part. Corporate colour palettes do not always
have a contrasting colour. In fact, often they have been
designed to work in harmony with one another. However
if you really want your call to action to ‘pop’ then you may
need to introduce a new colour into your arsenal.
The Things website draws attention to its calls to action
by using a contrasting colour
But contrast doesn’t just apply to colour. It also can
apply to size and movement too. Make sure your call to
action is substantially bigger than other links or buttons.
This makes it very clear to the user what to do next.
Equally, if the rest of your design is static consider
introducing some subtle animation to your call to action.
Please note the word subtle. User have become
desensitised to madly animating screen elements because
they are so heavily associated with advertising.
That said, if used in the right way, such as when there
is no other animation on the page, it can be effective.
So we have ensured our call to action is visible. We
have also made our message enticing. What happens
when the user clicks on it?
Following through on actions
Rarely is clicking on a call to action the end of the
process. Normally additional steps are required to
complete the purchase or fill in the contact form. Even
when no more steps are required it is important not to
abandon the user at this point. How you follow through is
Obviously how you respond will vary massively
depending on the call to action. However, I would like to
briefly share 6 guiding principles to help you follow
through on your calls to action:
‣ Track the action - For your own benefit make sure
you track any click on a call to action. This is
important not only in tracking the success of the
call to action. It is also important to ascertain
whether what happens immediately thereafter is
effective too. For example if the user clicks a
‘download our free ebook button’ and yet don’t
actually download the file, then you know there is a
problem with the screen following the click.
‣ Confirm the action - I have encountered websites
where clicking a call to action has led to no user
feedback. For example, I once completed a contact
us form and hit submit but nothing happened. Had
my email been sent? Had it failed? Make sure you
tell user what has just happened.
‣ Explain what will happen next - Once the user has
completed an action they need to know whether
any further steps are required and if not what
happens next. Will they receive an email? Will they
receive your product? What will happen?
‣ Continue the relationship - Calls to action can often
feel like the end of the relationship between you
and the user. Perhaps the user has placed an order.
Perhaps they have recommended you to a friend.
Whatever the case never let the relationship die.
Send them an email, add them to twitter. Just look
for away of staying in touch. As I have said before
repeat visitors are a valuable way to maximise ROI.
‣ Avoid surprises - Possibly the most important point
of all. The user should experience no surprises after
clicking on a call to action. That means no
unexpected delivery charges or that monthly
newsletter turning into daily sales pitches. This only
leads to unhappy users and we have already
discussed the dangers of that.
Wordpress do a great job at minimising the effort a user
has to put into creating a blog after responding to a call
To be honest somebody could write an entire book on
following up calls to action. However, the above advice is
a good starting point and will avoid losing the user at the
Hopefully, I have clearly demonstrated the importance of
calls to action on increasing your sites ROI. It’s probably
not much of a surprise. However, more importantly I have
outlined ways to identify your calls to action, ensure they
are effective and follow through when users respond.
Now for the important bit. Here are the actions I want
you to complete.
Action 1: Establish your main actions
Begin by writing a list of your sites main calls to action.
Next prioritise this list so your designer can ensure
users focus on the right actions first. Also consider how
one call to action can lead to the next.
Action 2: Carefully craft your actions
Once you have your main calls to action, work with a
designer and copywriter to ensure they are both
engaging and immediately obvious to users.
Action 3: Give every page an action
Make sure that your content authors add a call to
action to every page. Users should never be left at a
dead end so consider implementing the pagetables I
mentioned in this chapter.
By following the advice here you will increase the number
of users responding to calls to action and so boost your
ROI. However, there is always room for improvement.
The only way to work out how to improve things is to
release your site, test and iterate. It is this process of
refinement we will discuss in the next chapter.
A cycle of
In the Gifts of Sobriety, Barbara Cole wrote:
“We are locked into a cycle of repeating the same thing
over and over again, expecting different results. This is
commonly known as the definition of insanity.”
I have to confess this quote often springs to mind
when talking to website owners who have commissioned
yet another web designer to ‘redesign’ their site.
It is a pattern I see again and again. Websites are
launched to much celebration but over time they are
neglected to the point where the content is irrelevant,
while the design and technology is out of date.
Eventually, this leads to the organisation deciding that the
website needs to be ‘dealt with’ so they hire a web
design team to completely rebuild it. And so the cycle
Of course this works out nicely for us web designers
ensuring a constant stream of work. However, is it really a
good idea simply repeating the same process over and
over again, expecting different results?
I wouldn’t dare to suggest for a minute that all website
owners are insane. However, I do wonder if there is not a
1. Evolution not revolution
2. A different relationship
3. The sprint
4. An outside perspective
5. Multivariate testing
6. Next Actions
A cycle of reﬁnement 97
Evolution not revolution
In many ways this periodic redesign every few years is like
some central American coup or good old fashioned
revolution. Often it involves bringing in a new web design
team and throwing out everything that was done before.
The old order is swept away with little consideration
being given for what was good or bad about what went
It is understandable. Many organisations become
entirely disillusioned with their website and can no longer
see any good in it. They just want it gone and something
new and shiny put in its place. However, this is massively
wasteful. Why throw away perfectly good work? Instead
we should be building upon it to take our websites to
The evolution of the Amazon shopping cart is an excellent
example of incremental improvements.
After all, continually evolving our websites over time
brings with it considerable advantages.
98 A cycle of reﬁnement
The beneﬁts of evolution
There are good reasons why large online brands are
continually tweaking their websites rather than
redesigning them every few years. Take Amazon for
example. When was the last time you saw them do a
wholesale redesign? What about ebay, Google or any
other big player you care to mention?
Although Google has changed since its launch in 1998,
they haven’t gone through a major redesign.
Instead they have come to realise that a gradual
evolutionary approach provides a considerably better
return on investment.
So why does it provide better returns and does that
principle apply to you? I believe there are five reasons to
adopt an evolutionary approach to your website:
‣ More financially prudent
‣ Creates more buzz
‣ More engaging
‣ Doesn’t frighten users
‣ Is easier to test
However, it is probably the financial argument that is
the most compelling.
A cycle of reﬁnement 99
More ﬁnancially prudent
Redesigning a website is not cheap especially if the site is
big and you want to do the job well.
With big changes comes lots of uncertainty and that
means testing. Lots of testing. You need to test the
design, the content, the technology. It all needs to
checked and double checked.
Building on Then you need to do the design itself. When the
past versions design is going to be unchanged for 3 years everybody
of the wants to get it perfect. That means more testing,
website discussion and iterations.
rather than Once the design is approved it has to implemented,
reinventing the which involves updating every page on the site. This is no
wheel mean feat even presuming that you are not changing the
increases content, information architecture and technology too. If
your return you are, then you can add a couple of zeros to the price.
on investment. This is a massive capital expenditure to find every few
years and is rarely factored in to long term business plans.
It is money that has to be ‘found’ from somewhere.
Compare that to the evolutionary approach. No big
redesign, no change in technology, no massive rewrite
and restructuring of content. Instead you simply have an
ongoing series of tweaks and changes. Best of all you are
not throwing away previous financial investments. Instead
you are building upon them in an attempt to improve the
In effect you replace that one off capital expenditure
every few years with an ongoing investment in your
website. If you are sensible you will probably end up
spending about the same, perhaps even less. However, it
is a lot easier to budget for and because it is building on
the past rather than replacing it, you will generate better
However, that isn't the only benefit of evolutionary
design. There is also the PR opportunity.
100 A cycle of reﬁnement
Creates more buzz
From a marketing perspective, evolution offers some
exciting opportunities. With periodic redesigns you only
get one opportunity on the day of its launch to tell the
world about your wonderful new website. However with
evolution things are different.
With an evolutionary approach you have regular
opportunities to promote new features, improved
usability and other aspects of your evolving website. This
helps to create a continual background buzz and leave
others with the impression that your website is a hive of
activity and innovation.
Google is able to generate significant interest in the
incremental changes it makes to its website, such as the
launch of Google instant.
Best of all this doesn’t just attract new users in. It also
attracts existing customers back.
I have already talked about the importance of repeat
users, and an evolving website is a great way to increase
If you are constantly telling your existing users about
new features and improvements to your site you achieve
A cycle of reﬁnement 101
‣ One: You demonstrate a commitment to the site
and to your customers. So many websites are
effectively abandoned that this kind of commitment
makes you stand out from the crowd.
‣ Two: You have a reason to keep talking with users.
This helps to keep your brand in the front of their
‣ Three: You leave users with the sense that they are
missing out on something and this encourages
them to return to the site.
In short, it gives you an opportunity to encourage past
users to return to the website. However, evolution also
helps existing repeat visitors.
Doesn’t frighten users
The web is a complex place that is often hard to navigate.
Therefore when we find a way to get something done on
a website we stick to it. There maybe more efficient or
alternative ways to achieve the same thing but we stick
with what we know. We don’t like to change.
Digg.com experienced a dramatic reduction in traffic
shortly after the launch of a radical redesign.
102 A cycle of reﬁnement
Unfortunately, when website owners redesign their
sites the user is suddenly confronted with enormous
change that undermines the ways we have been working.
Even if the redesign is more efficient we are so
overwhelmed by the sudden change that we hate it.
We have seen several high profile examples of this in
the past. When Facebook redesigned it caused a huge
backlash from their community. The same is true for a
recent redesign of social news website digg.com who
have seen a dramatic fall in traffic since its redesign.
Substantial redesign’s risk alienating your existing users.
This is a dangerous area because you are alienating
your most passionate and valuable users.
Of course, with evolution you are still making changes
but they are on a much smaller scale and longer time
frame. It provides change for the user to adapt and
You can also be more confident that evolutionary
changes will actually be an improvement.
A cycle of reﬁnement 103
Is easier to test
I have already talked about the cost of testing a site
redesign. However, no matter how thorough your user
testing you can never be entirely confident until it goes
live. Real users will always respond differently than those
in a usability lab and that is hard to account for. Also,
when redesigning you are changing so much. Its often
hard to tell what is working well and what is not.
However an evolutionary approach has two
advantages when it comes to testing. First, it can be
tested on a segment of your live audience. In other
words. you can present a change to only some of your
users and see how they respond before rolling it out to
If the user Second, because you are only testing a single feature
or tweak it is possible to be much more sure of what
works and what doesn’t. When testing a whole redesign
redesign it is
there are just too many variables to account for. If the
hard to know
user rejects the design it is hard to know exactly what it is
about the redesign they are rejecting.
it is about
the design Hopefully by now you are seeing the benefits of an
evolutionary approach. There are, however, certain
processes that need putting in place before an
evolutionary approach can be adopted.
Laying the right foundation
“Hang on a minute” I hear you cry “are you saying you
never work on site redesigns at Headscape?” No, I am
not saying that at all. In fact, it is still the norm for us to
work on complete site redesigns. However, whenever we
do, we encourage clients to think of it as the last.
In order to evolve a website over time you need to
have a firm foundation. Unfortunately this foundation is
104 A cycle of reﬁnement
The problem is that the web is still relatively new. It
has been evolving at such a rapid pace that it has been
hard for any web designer to predict what building blocks
need to be in place for the future. Now though, things
are beginning to change. Best practice has emerged and
although there are still incredible innovations emerging it
is much easier to lay foundations that future development
can be built upon.
In fact I would argue there are only two factors that
need to be in place:
‣ A separation of content, design and functionality
‣ A flexible design
Let me explain what I mean.
Keep it separate
There was a time when the design, content and
functionality of a site were mixed together. This made it
hard to make changes to any one of the three without
having an impact on the others. Even something as
simple as changing a link colour could involve updating
every single page of your website.
Thankfully these problems are largely behind us and
most websites these days are written using modern
standards and content management systems.
However, things are not always as easy as they should
be. Sometimes web designers can be sloppy or things
have been implemented with speed in mind rather than
long term durability.
Sometimes content management systems do not help.
Although many websites have now been built with best
practice in mind their underlying content management
systems have evolved over years. As a result, they
continue to generate code that intertwines functionality,
design and content. This can prove problematic when you
wish to make a small change across an entire website.
A cycle of reﬁnement 105
However, the problem isn’t entirely a technical one. If
you want to evolve your website then you also need to
ensure the design is flexible enough to accommodate
Site wide changes can be expensive if the website is not
Ensure it is ﬂexible
If your website is to evolve, you will be constantly making
minor changes across the entire site. This will include
adding new content types and tweaking layout.
These kind of changes raise two issues. First, who
does them? Second, can the design accommodate them?
When I ask who does them, I am referring to whether
the website owner can (with a little training) make these
changes themselves or whether they have to return to the
106 A cycle of reﬁnement
As you will read later in this chapter I do believe in a
close, ongoing working relationship between website
owner and designer. However, I also recognise that going
back to your web designer will become expensive. If you
have to pay for every tiny change it is going to stop you
experimenting and implementing incremental
improvements. I therefore believe that whatever design is
implemented by the web designer should be flexible
enough to accommodate changes by the website owner
at a later date.
My second question is ‘can the design accommodate
changes?’ full stop. I come across many websites that
although beautiful are so hand crafted that change
involves substantial reworking. For a site that needs to
focus on ROI through constant refinement this can prove
With this in mind I believe that websites focused on
ROI need to work within these two constraints. The
design is flexible enough to adapt overtime and the
website owner has at least some ability to make updates
How then is this possible? The approach that we are
increasingly taking at Headscape is modular design. By
that I mean we create an underlying grid system for the
entire website and then build a series of design modules
(elements such as navigation, news listings, and related
links) that can be dropped into this grid.
The clever part is that instead of just designing these
modules to work in a single location (e.g. The narrow
right hand column) they can also be used in any other
position and the design automatically updates to work in
For example a list of news stories might include an
image when displayed in the main content area because
there is ample space. However when placed in a narrow
column the images might disappear to work with the
A cycle of reﬁnement 107
Headscape used a
modular design on
the OU website to
give the client
108 A cycle of reﬁnement
This kind of adaptive design enables the website
owner to test out different positioning of elements
without the intervention of a designer and modules can
be recombined in new combinations for further templates
that need producing.
That is not to say all modules will work perfectly in
every combination. Certain combinations may draw the
users eye to the wrong part of the page or distract from
the primary call to action. That is why it is still good to
have an ongoing relationship with your web designer so
you can get their opinion on changes being made. This
brings me on to the next secret of successful websites; a
different kind of working relationship with your web
A different relationship
There is a problem in the way web designers and website Don’t miss
owners work together. It is a problem that leads to less out on your
effective websites and website owners receiving less web designers
value for money from their web design teams. primary
The problem lies in the constant cycle of redesign I strengths;
have already written about. In such a system the his knowledge
relationship between web designer and website owner is of the web,
limited. The website owner decides what they want to what others
build and the web designer builds to that specification. are doing and
This all sounds perfectly logical however it means that best practice.
the website owner is missing out on one of the web
designers primary strengths; his knowledge of the web,
what others are doing and best practice.
Don’t waste your web designers
For many years in my web design career I would receive
briefs from clients and simply respond saying exactly how
much it would cost to implement their ideas. Often I
A cycle of reﬁnement 109
disagreed with these ideas. In many cases I knew of a
better approach. I always wanted to know more about the
motivations behind the clients ideas and to work in
partnership with them to find the best solutions.
However, the relationship didn’t enable that.
This kind of approach does not get the best returns from
your web designer.
These days I have a very different relationship with
most of my clients. I am more confident to challenge the
brief I am given and suggest alternative ways of solving
the clients problems. I also ensure there is always a
requirement gather phase where I get to understand the
business objectives and constraints.
This is a massive improvement for all involved. My
clients get the full value of my knowledge and skills
110 A cycle of reﬁnement
(rather than just a subset) and I am perceived as adding
value added services that differentiate me from many of
Even this is not the perfect solution. It falls down
because the invitation to tender still frames the scope of
the conversation. It is still the website owner who defines
what issues need addressing even if I now get to
influence how these problems are solved. I do not often
get to actually recommend what issues should be
addressed and define the priorities for future
development of the website.
There are some exceptions. I work with some
particularly savvy clients who understand if they include
me at the start of the process they get more value from
me and end up with a better website.
So what am I suggesting? What is the alternative?
By working collaboratively with their web design agency
Wiltshire Farm Foods have increased profits by 10,000%
over 8 years.
A cycle of reﬁnement 111
A better way
Instead of you creating a detailed brief covering
everything you want a web designer to build, why not get
the web designer to do it for you?
Beginning by commissioning a web design agency to
do a review of your website. Outline any problems you
perceive with the site and give the web designer some
guidance in terms of your business objectives. Don’t
define what solutions you wants to these problems. Allow
the web designer to review the site and make
suggestions about how it could be improved to better
meet your business objectives.
The deliverable for this first “micro” project would be
a report outlining all of the things that the web design
agency suggest you do to improve your website.
Begin the Traditionally, employing a web designer has been a
relationship huge risk. Because it cost so much to redesign a website
with a new it is important you get the right agency. Failure to do so
can become costly. However with this new approach you
have the opportunity to get to know the web design
agency by using them on a very small self-contained
them to do a
project (the site review). This provides you with the
opportunity to assess them and decide if you wish to
This allows continue with them for the main build. If for any reason
you to you are unhappy there is nothing to stop you taking their
assess their report to another web designer to implement.
Even after this initial report has been delivered, I
would still encourage you to use a “micro” project
perspective approach. In other words, instead of implementing all of
on your site. the suggestions made by the web designer in his report,
split the work into a series of smaller projects.
Again, this allows you to limit the risk associated with
bigger projects and continue to build a relationship of
trust with your web design agency. If at any stage you’re
unhappy with the work being produced there is nothing
112 A cycle of reﬁnement
to stop you moving to a different web designer for the
next small project. These small projects are often referred
to as “sprints”.
By now you’ve probably gathered I’m not a fan of large I am not a
website redesign projects. They are expensive, risky, and fan of large
prone to slippage. They also fail to properly engage the website
web designer in the process. redesign
Instead, I believe that the most successful websites projects.
come about because the website owner and web They are
designer work on small clearly defined projects. Each expensive,
project is completed in a short period of time (called a risky and
sprint) that ranges between a week and month long. prone to
What is a sprint?
To give you an idea of what a sprint might consist of, I
have provided a few examples below. Notice that each
sprint focuses on one key element and at no stage
involves redesigning large areas of the website, let alone
the whole thing! Some possible sprints include:
‣ Testing a number of different calls to action to find
out which one is most effective.
‣ Implementing a new piece of functionality such as a
contact look up application.
‣ Making adjustments to the look and feel of the site
to improve accessibility.
‣ Changing the sites information architecture to be
more focused on user needs.
‣ Rewriting and reorganising key pages to better
reflect the evolving priorities of the company.
Obviously this list is just the tip of the iceberg but it
should give you an idea of what I mean when I talk about
a sprint or a small "micro" project.
A cycle of reﬁnement 113
To get the best return on investment from your sprints
it is important to run them effectively.
Running an effective sprint
Between yourself and the web designer you should have
a pretty good idea of how your website can be improved.
The next step is to plan a schedule of development to
address the various problems you have identified. As I
have already suggested the best approaches to break it
down into a series of small sprints.
Although the length of the sprints can vary, I believe a
good place to begin is with month-long sprints. This ties
in well with the monthly cycle of usability testing we have
As I've said before I recommend holding monthly
conference calls with your web designer to establish what
work will be undertaken each month. This should include
a mixture of new features that you wish to develop your
website, solutions to problems you have identified and
issues that have come out of your usability testing.
To reduce costs consider a service like GoToMeeting.
This will allow you to discuss work with your web
114 A cycle of reﬁnement
Each call should also discuss progress made since the
last meeting and whose responsibility it is to do what
work. For example, if budgets are tight then you may
choose to do more of the work in-house rather than
outsourcing to your web designer.
However, whether you are doing development in
house or have chosen to outsource it, I still recommend
including your web designer in your monthly review
process. This call only needs to be 30 minutes long and
so won't prove expensive. The reason I make this
recommendation is because I believe it is extremely
important to have an outside perspective on your
An outside perspective
So far I have focused on website owners’ relationship with Even if you
external web designers. Increasingly many organisations have an in-
have internal web teams with whom there is already a house web
collaborative relationship. team it is
The principles I have outlined so far apply whether you still worth
are using an internal or external web designer. However, I employing an
would argue that even if you have an in-house web team external
it is still worth including an external viewpoint in your agency to
monthly sessions. provide an
Admittedly, as a part of an agency myself you will outside
imagine there is a certain amount of bias on my behalf perspective.
and you might be right. However, I confidently believe
that getting an outside agency involved will dramatically
improve your website even if they do not do any of the
implementation work and merely consult. Why?
There are several problems with relying solely on your
internal web team:
‣ Because they are so entrenched in your
organisational structure it is hard to remain
A cycle of reﬁnement 115
objective about what users will easily understand.
What is obvious to them may not be to the user.
‣ Being so involved in the day to day development of
the website often makes it hard for internal web
teams to see the broader perspective. They
become overwhelmed by details and lose the
‣ In my experience internal web teams are almost
always overworked. The ongoing maintenance and
development of the website leaves little time for
considering broader issues like the future roadmap
of the site. They often become reactionary instead
of developing strategically.
‣ Because your internal web team is working on a
single set of websites they lack the broader
perspective than an outside web designer has.
These individuals work on a huge variety of sites
and can bring that experience to bear when
looking at your site. I am often taking what I learnt
on one website and applying it to another.
‣ Finally by their nature most internal web teams
have to be made up of generalists. From time to
time therefore issues arise that are more
specialised than the internal team can handle. In
such cases having access to an external team of
specialists is useful.
I am not suggesting that having an internal team is a
bad idea or that external web designers are better. I am
just saying an outside point of view will bring a new
perspective to your site.
This doesn't need to prove expensive either. I am not
suggesting all work identified in the monthly review
needs to be outsourced. If you already have an internal
team everything except the most specialist of work may
well be done in-house (workload allowing). The outside
web designer may do nothing more than attend your
monthly review meetings.
116 A cycle of reﬁnement
Working together with your web designer you should
be able to identify ways that your website can be more
profitable and increase your return on investment.
However, you can never be 100% sure your improvements
will help. That is why if you want to really enhance your
websites effectiveness you need to test.
So far we have talked about user testing, which although
powerful does have some limitations. The biggest
limitation is the fact that any user testing takes place in a
false environment. Users are observed and asked to "talk
out loud". Both of these behaviours are not natural.
Fortunately, there is a way you can test that completely Multivariate
replicates a real experience. This is multivariate testing.
Multivariate testing involves testing different elements different
of a user interface on your live site by presenting elements of a
alternatives to a small subset of users. interface on
For example, let's say you have concerns about a call your live site
to action. The current solution is not performing as well as by presenting
you would hope and so you decided to test some alternatives
alternatives. Using multivariate testing you would create to a small
several alternative calls to action and then deliver these subset of
alternatives to a small number of users on the site. The users.
remaining users will continue to receive the original call to
action until you have identified the best solution.
By displaying the alternative calls to action to a subset
of your users you can identify which one performs best
(which users click on the most) and then roll this solution
out to all users across the website.
Multivariate testing is a powerful tool that can be used
to test a variety of different issues. In fact with a little
thought you can test almost anything.
A cycle of reﬁnement 117
What to test?
Unsurprisingly multivariate testing is most commonly used
on transactional websites. Small changes on these sites
can make a big difference in revenue especially on busy
websites. Multivariate testing can be used to test:
‣ The checkout process.
‣ Messaging to users about security.
‣ How users respond to different discounts.
‣ The placement and design of calls to action.
‣ User registration and login.
The list goes on. However, multivariate testing is not
limited to transactional sites. It could also be used to test:
‣ A signup form on a web app.
‣ The most effective headline on a blog post.
‣ Alternative copy on a call to action button.
‣ Different labels on site navigation.
Huffington Post is just one news website that uses
multivariate testing to find the best title for stories.
118 A cycle of reﬁnement
In short it is possible to improve the performance of
almost anything on your website. It is even possible to
test completely different versions of a design or individual
page. All that is required is some method to measure
which version users prefer. This could be as simple as
which option the user clicks more or as complex as which
option eventually leads to the most sales.
However, don’t just test elements which are
performing badly. You can also use multivariate testing to
remove doubt and overcome disagreement. How many
times have you had disagreements within your web team
over the best way to solve a problem? I am guessing a
lot. In such situation test the alternatives. This is much
more effective than arguing over personal opinion.
There is little doubt that this is by far the most
effective way of optimising your site to meet both your
users needs and business objectives. Best of all there are
free tools available that make multivariate testing
incredibly straightforward to set up.
Google Website Optimiser provides a free introduction
into the world of multivariate testing.
A cycle of reﬁnement 119
How to test?
Multivariate Although there are a variety of applications available to
testing is help you do multivariate testing, I would suggest the best
about testing place to begin is by using Google website optimiser.
individual Not only does this integrate well with Google analytics
elements (which we discussed earlier) it is also free and easy to use.
within a page. A good place to begin is by carrying out simple A/B
A/B testing test. Google’s A/B testing is similar to multivariate
allows for testing, but instead of testing multiple elements on a
the testing of page, you test entire pages. This is ideal for website
a number of owners just beginning to explore multivariate testing. It is
different also good for websites that do not have a high enough
versions of a level of traffic to support multivariate testing. Best of all
single page. with Google website optimiser you can set up a basic A/B
test in less than 5 minutes.
Google provides an excellent tutorial video showing
you how to set up a simple A/B test. Essentially this
consists of three very simple steps. Step one consists of
creating various versions of the page you wish to test and
adding each of these to Google website optimiser. In
step two you need to identify the action you wish users to
complete. This could involve:
‣ clicking a link,
‣ pressing a button,
‣ submitting a form,
‣ or visiting a certain page.
Once you’ve completed step two and told Google
website optimiser what your conversion point is, all that
remains is to copy and paste code provided by Google
into each variation of your page and associate a small
piece of code with the final action you wish people to
Once you have made your test live, different users will
see different versions of your test page. If you’re worried
about too many users seeing the experimental page you
120 A cycle of reﬁnement
are testing, it is possible to limit the percentage of users
seeing this alternative page. You can even configure the
website optimiser to automatically disable poorer
performing versions of your test page. This ensures the
maximum number of users possible are seeing the best
Setting up an A/B test really is as simple as that. There
is no reason why you cannot do it yourself if you have
access to the HTML files on your website. If not, Google
provides a link you can send to your web designer telling
them exactly what to do.
But why stop with a simple A/B test? There is also an
option to do much more complex multivariate testing.
However even this is not as intimidating as it sounds.
If you have some budget available then check out Visual
Website Optimizer. It makes the process of creating
multivariate testing even easier.
The advantage that multivariate testing provides is
the ability to test multiple elements within a single page.
You could test three different headings alongside two
A cycle of reﬁnement 121
different images and four different calls to action. This
would enable you to identify which combination of
heading, images and calls to action worked best together.
Setting up multivariate testing is similar to A/B
testing. The only difference is that instead of creating a
series of completely separate pages you instead
identifying areas within a page that will change. Then
within Google website optimiser you create a number of
alternative versions of these different page elements.
Once you’ve done that you are ready to go.
Whether you are using A/B or multivariate testing,
there is little doubt that the return on investment is
significant. Instead of relying on personal opinion or
limited user testing, multivariate testing ensures that you
have the most affected website possible. This is especially
true with an ongoing programme of testing that
continually tweaks and improves your site. This is yet one
more reason why I recommend an ongoing policy of site
development rather than sporadic redesign every few
So what is stopping you? Why not try out multivariate
In this chapter I have focused on the need to evolve your
website rather than redesigning every few years. I’ve
explained that to achieve this website owners need a
different kind of relationship with their designer. A
relationship where both parties work together in a series
of ongoing sprints.
In order to achieve this I suggest the following actions:
Action 1: Establish the right relationship
Establish a new working relationship with your web
team both in terms of contractual arrangement and
ensuring you work as a partnership. Look for a web
designer who can be your long-term partner rather
than simply a short-term supplier.
Action 2: Schedule monthly meetings
Make sure you are discussing the future of your website
at least once a month with your design team. Plan what
work you are going to complete over the coming
month and spend at least some time considering the
long-term strategy of your website.
Action 3: Arrange regular testing
Ensure that testing is at the heart of any process of
development used to improve your website. This
should be a combination of user testing and the
multivariate testing we’ve discussed in this chapter.
So far this book has focused on the return on investment
we can generate from existing users. However, to create a
website truly focused on ROI you also need to consider
how to drive more users to your site.
In my book ‘The Website Owner’s Manual’ I begin the
chapter on driving traffic by referring to the Kevin
Costner film Field of Dreams. In that movie Kevin Costner
hears a voice that tells him, “if you build it, they will
come.” I suggested that many website owners had this
mentality believing that users would just turn up at their
site once it had been launched.
I think things have moved on since I wrote those
words. Generally speaking, I believe most website owners
now realise that their website needs promotion. However,
I believe there is still a lot of confusion about how best to
drive traffic to a website.
In this chapter I examine the most cost-effective ways
of driving quality visitors to your site. In particular I look at
search engine optimisation and social media. I want to
ask whether these two massively hyped approaches
provide the return on investment that most website
Before I can do this I think it is important to stress that
every website owner has different requirements and that
will affect which marketing approach works best.
1. Identifying your requirements
2. Tracking quality and quantity
3. Does SEO really pay?
4. Viral marketing and social media
5. Where to start
6. Next Actions
Driving trafﬁc 125
Identifying your requirements
The best When we talk about driving traffic to our websites it is
approach to important to realise this is not a black or white discussion.
marketing Different websites require different things from their
your site online marketing strategy.
depends on For example, start-up websites often have to focus on
target brand building and gaining market share. A mature
audience, type website may be more interested in increasing the quality
of site and of visitor and the number who go on to convert.
sector. The type of target audience also makes a difference to
the most effective form of marketing. Some audiences are
more likely to come to your site via a search engine while
others will be more open to recommendations from
friends (social media).
The sector you work is also a consideration. Some
sectors are very competitive and so paying for advertising
will probably be expensive.
With this in mind it is important for you to have a clear
understanding of what requirements you have from online
marketing campaigns. Without understanding who you
wish to reach and what you want them to do, it is going
to be impossible to track the quality of return you get
from your investment. I recommend asking the following
‣ Who are your target audience and how do they go
about finding products and services online?
‣ What do you wish to achieve from your marketing
campaign? Are you trying to build brand
awareness, increase dwell time, or convert users?
126 Driving trafﬁc
‣ How competitive is your marketplace?
‣ Are your competitors already paying for advertising
either on search engines or websites your
customers might be reading?
‣ How much can you afford to spend on promoting
Admittedly this last one is a difficult question to answer
especially when you have no frame of reference about
how much online marketing costs.
Look at it another way. Ask yourself how much each
new user to your website is worth to you? If you run an e-
commerce site this should be relatively easy to calculate.
It is simply a matter of calculating the average profit
made on each user who buys divided by the ratio of users
who actually convert. It is obvious, though, that not all
websites are pure e-commerce sites.
If your website is designed to generate leads, things
are slightly more tricky. However if you followed my
advice in chapter 2 (Measuring your successes) it should
still be possible to work out the ratio of users who convert
and the average profit made from each user.
For purely informational sites, it is still important to
establish how much you are willing to pay for each user.
Driving trafﬁc 127
This figure should include the cost of maintaining the
website as well as the cost of attracting the user in the
first place. Without knowing how much your website is
costing per user it is easy for costs to spiral out of control.
Recently in the UK this has become a particularly hot
topic with the government closing a number of websites
based on the exorbitant costs incurred to cater for each
Once you know how much you are willing to pay per
user and understand the other requirements surrounding
your marketing strategy it is possible to start
experimenting with a number of different marketing
tools. In order to identify the most cost-effective tool you
need the ability to be able to track the successes or
otherwise of each individual tool.
Tracking quality and quantity
It is not
enough to In chapter 2 (Measuring successes), I focused primarily on
measuring the success or otherwise of your website but it
is not enough to just track how effective your website is.
many people a
You also need to know whether the marketing methods
you’re using to drive traffic to your site are working.
drives to your
website. You To ascertain the most successful marketing technique
need to know for your particular website you need to look at both the
quantity of visitors being generated and also the quality
of those users in terms of conversion.
Why quality matters
Unfortunately, too many website owners focus exclusively
on quantity rather than the quality of visitor to their
website. The problem with focusing on quantity is that
although it is relatively easy to drive large amounts of
traffic to your website, it is much harder to get them to
stay and eventually convert.
128 Driving trafﬁc
Take for example the “Digg effect”. This phenomenon
is associated with the social news website Digg.com.
Many website owners work hard to produce content that
gets promoted to the home page of Digg.com. This is
because appearing on the homepage of Digg will drive
huge amounts of traffic to your website.
Unfortunately, there is a problem associated with
Digg.com referrals. The users coming from Digg are only
interested in the immediate piece of content promoted
on the Digg.com homepage. They will read that single
piece of content and then leave. Essentially those users
belong to Digg.com and not to you.
More traffic does not always lead to a higher return on
Although it is obviously possible to encourage some
of these users to stay the ratio that actually do so will be
low. Worse still, the sudden arrival of such large amounts
of traffic can actually be detrimental to the experience of
your existing users. For example it is not unusual for the
Digg effect to cause entire websites to collapse under the
weight of traffic.
Driving trafﬁc 129
Of course this problem is not just limited to Digg.com.
There are a number of other similar sites that many
website owners aspire to appear on. This focus on
quantity at the expense of quality is rarely an advisable
approach. How then can you be sure that the marketing
approaches you are using encourage quality traffic rather
than just quantity?
An introduction to tracking
It is Before I explain how to track the quality of leads coming
important to to your website, let’s first begin by looking at how to
experiment track users coming from different marketing channels.
and track Let’s say you are experimenting with three different
various approaches to marketing your website:
marketing ‣ You have purchased a banner advertisement on a
channels. website you know your target audience read.
‣ You have been participating in different forums that
your target audience frequents.
‣ You have commissioned a search engine
optimisation company to improve your search
rankings on certain keywords.
How would you track the success or otherwise of these
different marketing approaches?
Tracking banner advertising
Tracking banner advertising is probably the easiest of
three options. This can be done remarkably easily from
within Google analytics using its built-in campaign
By using Google’s URL builder you can create a
custom URL that will track users who click on your banner
ad. This allows you to see exactly which user arrived from
which banner ad in Google Analytics. You can also see
combined results for all banner adverts you are running
130 Driving trafﬁc
on various websites or detailed results for a particular
advert running on a particular website.
Google Analytics allow you to track the progress of
various marketing campaigns.
Once you are able to track users who have clicked on
your banner advert it is a straightforward matter to
calculate how much you are paying per user. But what
about something more complicated like tracking the cost
of users who’ve come in via social media?
Tracking forum participation
Many online marketeers suggest that participating in
virtual communities can be effective way of driving traffic
to your website. One example of this is participating in
forums that are frequented by your target audience.
However to track the success or otherwise of this
marketing approach you need to be able to identify users
who’ve arrived from the forums you are participating in.
This can be done in exactly the same way as banner
adverts. Using Google URL builder you can create a
unique URL that is added to your posts in each of the
forums you participate in. These will then appear in your
Google analytics results.
Driving trafﬁc 131
The tricky part is working out how much each of these
users are costing you. The only way to be sure is to track
the number of hours you spend contributing to these
various forums. If you can associate a cost to your time
then it is possible to identify how much each user costs.
2 hour spent in 20 visitors came from Profit = £80
the forum at £60 the forum of which 2 (£200 - £120)
per hour = £120 went on to place
£200 worth of orders
By monitoring your time in forums and the purchases
made you can establish whether this is an effective form
However remember that sales can take a long time to
come from forum contributions. Do not judge over too
short a period.
Unfortunately things become even more complex
when looking at the subject of search engine
Tracking search engine optimisation
In my example I suggested you had commissioned a
search engine optimisation company to improve your
search rankings on certain keywords.
Many companies pay search engine specialists a large
amount of money to improve their rankings. How can you
tell if this money is well spent?
Google analytics does allow you to view how many
users have arrived after searching on a particular keyword
or phrase. It is easy enough to divide the fee charged by
132 Driving trafﬁc
your SEO company by the number of users who arrived
via a certain keyword to get an indication of the price per
Unfortunately, things are never quite that simple. What
this model does not take into account are users you
would have received even if the search engine
optimisation company was not involved. You therefore
require a baseline in order to see the increase that the
SEO company has generated. In other words you need to
divide the fee charged by the SEO company by the
number of additional users who arrived by a particular
keyword or phrase.
Spam email promises us better ranking on Google
but does that improved ranking generate enough
revenue to cover the cost of employing the SEO
This combination of built-in Google analytics reporting
combined with custom URLs and enables you to track
how many users are coming from different sources.
However, this is only half the story. You also need to be
able to track how many of those users convert. This gives
Driving trafﬁc 133
a much better indication of the quality of different
Once you have the ability to track users arriving through
different marketing methods it is relatively
straightforward to track them to the point of conversion. I
say “relatively” because if that conversion point is off-line
you will need to put processes in place to make this
happen. For example if the call to action on your website
is to contact you by phone then you will have to actually
ask a caller how they heard about you. This is because
there is no way of tracking and individual campaign
without having a separate telephone number for each
It is important to ask whether enquiries have come via
the website to enable effective tracking.
In most situations it is possible to track conversion (or
at least the completion of a call to action) entirely through
Google analytics. As long as the conversion point involves
clicking a link, pressing a button or submitting a form
134 Driving trafﬁc
then Google analytics can track it. This enables you to
follow the user all the way from the original marketing
campaign to the point of conversion.
It is by focusing on which marketing campaigns
produce the most conversions that you gain a better
picture of what works and what doesn’t. It is through this
approach that you would identify problems with sites like
Digg.com that generates large amounts of traffic but low
levels of conversion.
Once you have the ability to track the quality and
quantity of leads, you now have the ability to find the
most effective marketing channel for you. However this
does rely on a considerable amount of trial and error,
which can be both expensive and time-consuming.
In order to minimise the amount of guesswork, I want
to look at two areas that seemed to be surrounded by a
lot of misunderstanding. These are search engine
optimisation and the use of social media.
Let’s begin by looking at the subject of search engine
For the vast
Does SEO really pay? majority of
There are no shortage of companies promising to owners an
improve your search engine rankings and even provide SEO company
the number one spot on Google. But can they really is not going
deliver what they offer? And even if they can will the cost to create a
of getting there generate a reasonable return on significant
investment. return on
To answer this question it is important to understand investment.
one important aspect of search engines. They are only
interested in one thing: providing relevant results to their
users. If your content is not relevant, no amount of clever
search engine optimisation will get you to number one.
Driving trafﬁc 135
There are ways to trick Google into ranking your site
better, but these break Google’s guidelines. Some search
engine optimisation companies use these techniques to
improve rankings. However in almost all cases, the benefit
is temporary, because Google works to close such
loopholes. What is more, Google comes down hard on
sites that disregard the guidelines.
BMW was famously removed from Google for using
unsanctioned techniques for improving their ranking.
That said there are legitimate techniques available that
will improve your search engine rankings. These focus on
removing obstacles that can hinder search engines from
indexing the content of your site. Before you hire a search
engine optimisation specialist be sure they are using
legitimate techniques (something I cover in more depth in
the Website Owners Manual) and ensure you have a way
of monitoring how much additional traffic your specialist
is actually providing.
136 Driving trafﬁc
In my opinion if you have hired a good web design
agency you shouldn’t need the services of an SEO
specialist. That is because your web designer should have
already removed any barriers that prevent a search
engine from indexing your website.
In my experience, many SEO specialists spend a lot of
time implementing detailed changes that fail to provide
massive returns. As a result SEO companies do not
generally generate enough additional income to justify
their fees. However there are one or two exceptions to
this rule. These are:
‣ Websites that generate a large amount of traffic.
In these situations a small percentage increase in
users from an SEO improvement will generate
‣ Websites that sell high-value products or services.
Improvements in search engine placement may
only generate one or two additional leads but
because of the high profit margin on the sales the
return on investment makes it worthwhile.
‣ Websites with a high conversion rate. If your
website converts a significant number of users who
arrive from search engines then the return on
investment from employing an SEO specialist may
‣ Websites that have been badly implemented by
web design teams who have ignored the basic rules
provided by Google for optimising websites.
However, if your website does not fall into the above
categories I believe you are much better off hiring a web
designer who has a good understanding of building
websites with search engines in mind than employing a
specialist in SEO.
In short, there are a lot of people working in the SEO
field who are making promises and offering services
which are just not right for the majority of website
Driving trafﬁc 137
However, this is not just an issue in SEO. It also applies
to social media.
Viral marketing and social media
It is hard to turn around on Twitter these days without
running into a social media guru. It would seem that
anybody who knows how to set up a Twitter account or
Facebook page can now call themselves an expert in
There is no doubt that social media is massively overhyped
at the moment.
A consequence of this glut of so-called experts is that
many are making unrealistic promises about what social
media and viral marketing can do.
Don’t get me wrong, I believe that online communities
and word-of-mouth recommendation can be some of the
most powerful forms of marketing available to us. I also
recognise that it does not always generate the kind of
return on investment many promise. Neither is it the
quick win that it is often portrayed as.
So what are the supposed benefits of both social
media and viral marketing?
138 Driving trafﬁc
The beneﬁts of social media
It would be hypocritical of me to suggest that social
media has no value. My own reputation and the new
business generated for my web design company
Headscape, comes almost exclusively through the use of
social media. Every day I use twitter, Facebook,
podcasting and blogging as a way of reaching potential
customers. The fact that you are reading this book is a
testament to the fact that social media works.
I use social media all the time to communicate with
potential customers. The fact that you are reading this
book is testament to the fact it works.
Social media is not just a way of directly reaching
customers. In chapter 3 (Becoming user focused), I
demonstrated that building an active community using
social media tools can create advocates who go out and
promote your products and services on your behalf. Any
marketeer will tell you that this kind of word-of-mouth
recommendation is invaluable.
Although I am the first to champion social media as a
marketing tool and as a way of engaging with your
customers, it is not without its limitations.
Driving trafﬁc 139
So what’s the problem with social media?
Unfortunately I do believe that social media should be part of every
social media organisation’s attempt to drive more traffic to their
has been so website and convert more users. However the main thrust
overhyped and of this book is about return on investment. Although I do
little has believe social media provides a good return on
been written investment is not something that happens overnight.
about this Social media has been so overhyped and little has
huge been written about this huge time and resource
commitment commitment it requires. Unless you choose to hire a social
it requires to media expert it won’t necessarily cost you a lot of money.
make happen. However, it does require an ongoing commitment.
The key to a successful social media strategy is two
fold. First get exposure and then build relationships.
Before you can expect user to start promoting your
products and services, they first need to have heard of
them. This is where exposure comes in. In the early days
of using social media you need to be everywhere. You
need to be participating in forums, posting on twitter,
writing blog posts and engaging on Facebook. Before a
user is going to be curious about what you have to offer,
they need to see your name pop up again and again. This
requires a lot of hard work and continual commitment. It
is not something that will happen overnight.
Once potential customers have started to show an
interest by visiting your website, subscribing to your blog
or following you on Facebook or Twitter, the next step is
to build relationship with them.
Unless a user likes you and feels you are sincerely
interested in hearing what they have to say, they will not
become an evangelist for your product or service. This
means that you have to invest time in building the
relationship to a point where they consider your website
as important to them as it is to you. Only then will they
start promoting it on your behalf.
140 Driving trafﬁc
The lesson to be learned here is that although social
media can be incredibly powerful it can also be expensive
(if you use an outside expert) or time-consuming (if you
do it yourself). Worse still, if you do not commit to it for
the long term it is easy to stop before you reap the
returns on investment it can provide.
I would encourage you to include some element of
social media in your marketing strategy. However, be
aware that this is probably going to require a couple of
hours per day of somebody’s time for several months
before you begin seeing return on investment.
And social media is not the only overhyped version of
word-of-mouth advertising. There is also viral marketing.
What about viral marketing?
We have all seen the stories of overnight sensations on
YouTube that generate millions of views in a matter of
days. We have also seen links to amusing content, useful
websites or the latest fad circulate like a virus.
Old Spice is just one example of a phenomenal viral
sensation that was viewed by millions.
Driving trafﬁc 141
Viral In the early days of social media these kinds of viral
marketing is phenomenon were largely produced by enthusiastic
primarily individuals. Increasingly they are actually now being
suited to carefully crafted by advertising agencies. At the time of
large brands writing this the latest viral sensation has been the Old
looking to Spice advertising campaign. These high quality, amusing
videos have swept the web at an extraordinary speed and
have no doubt been an incredible boost to the Old Spice
However does that mean we should all be looking at
trying to create the next viral sensation? I would argue
that unless you are an major brand trying to target a large
general public audience this is probably not the most
effective kind of advertising for you.
I believe there are three reasons why viral marketing is
not particularly good tool for generating return on
‣ It is much harder to target a specific audience. With
viral marketing, you have little control over who is
exposed to your content. Although the nature of
the content has some influence, viral marketing
isn't as refined as other methods.
‣ Those who view your viral marketing campaign do
not care about your site. Instead they are
interested in the content of your viral marketing,
not what your site can offer. This significantly
‣ They are unlikely to remember you. The typical user
who views viral marketing sees a lot of similar
content. Unless yours is truly exceptional, they are
unlikely to remember you. Even if they do, they're
more likely to remember the marketing rather than
I am very aware that so far in this chapter I have
focused on the limitations of various marketing
approaches. I've discouraged you from using search
engine optimisation companies. I suggested that social
142 Driving trafﬁc
media (although very effective) won't provide quick wins
and that viral marketing is only for large consumer
brands. Where then should you start in creating an
effective marketing campaign for your site that will
generate return on investment?
Where to start
I do believe that search engine optimisation and social
media have their benefits and should be part of your
marketing strategy. The problem is not so much in the
various techniques available as in having unrealistic
expectations about the kind of return on investment you
I think the biggest problem for many website owners
is that they do not have a strategic plan in place for
promoting their website. The first step in any such plan is
to ring fence time and money for marketing.
Ring fence time and money
Who is responsible for promoting your website? Many Who is
organisations couldn't answer that question. Even if there responsible
is a person responsible for marketing the site, often this is for promoting
just an unofficial add-on to their normal day job. As a your website
result it never gets the time and attention it really needs. online? If
Most organisations recognise the need to invest in their you don’t
website but not really in promoting it. know its time
Therefore the first step in any successful marketing to decide.
initiative is to ring fence somebody's time to work on
marketing. This means writing it into their job description
and identifying how many hours a week they're going to
spend on the task.
In some cases setting aside time alone is enough. If
your main marketing is through Google Adwords or
banner advertising you will need to consider an ongoing
Driving trafﬁc 143
budget to support these online marketing activities. This
budget should be based on the amount you are willing to
pay per new user as we discussed earlier.
Deciding on whether you need money set aside for
online marketing is very much dependent on who your
audience is and where they can be found online.
Finding your audience
If you want to generate the greatest return on investment
for your marketing efforts it is important that those efforts
are targeted at your specific audience. This means
understanding exactly where that audience can be found
online so that you don't waste time communicating with
people who will not buy.
Depending on your type of client finding your target
audience can be as easy as asking existing customers.
Ask your customers where they hang out online. This
will help target your marketing.
When Headscape decided that it wished to target the
higher education sector we asked one of our higher
144 Driving trafﬁc
education clients what forums they used, what blogs they
read and what mailing list they participated in. This gave
us a clear indication of where we should focus our
marketing messages. It also informed us which marketing
tools would be most effective. Because of the way this
community of users interacted with one another we were
much better off using social media tools rather than
relying on paid for advertising.
Unlike traditional advertising on television or radio it is
actually possible to be very specific about who you are
talking to online. Let's say that you wanted to reach pre-
teen girls in London. You could easily use Google
Adwords which allows you to both target by geography
and keywords. Alternatively, you might wish to consider
Facebook advertising that provides incredibly specific
targeting of your adverts.
Facebook allows you to be incredibly accurate in who
you target with your advertising.
Driving trafﬁc 145
For a more general audience, such as the over 60s in
the UK, you can participate in the many forums aimed at
this age group or write posts for blogs who focus on this
However be warned! If you are going to participate in
forums or write posts ensure they are more than glorified
Don’t just advertise, provide value
If handled Although participating in forums, chat rooms, mailing lists
badly or other forms of online community can be an effective
participating way of driving traffic to your site, it can also be damaging
in online to your brand.
communities The problem is that increasingly website owners are
can damage recognising the marketing value of social sites and
your brand. spamming them with blatant advertising. This only serves
to anger users and damage the perception of your brand.
To effectively market on these social sites, a more subtle
approach is required. Always adhere to the following
‣ Build a reputation. Never just start shouting about
your site. You must earn the right to promote it.
People need to know and trust you. This involves a
long-term commitment to that community.
‣ Use your profile. Most communities provide profile
pages. These allow you to post basic information
about yourself and link back to your site.
Completing your profile shows other users you
committed to the community and also provides a
‣ Don't self promote. Never post about your site
directly. Instead add a small link at the bottom of
your post in the form of signature.
‣ Follow the rules. Be sure to read and follow any
community rules about self-promotion, or risk
146 Driving trafﬁc
‣ Contribute value. Avoid adding posts with no
value. Posting “yeah, I agree" or “good point"
followed by a link to your site is spam. Build
respect by posting useful responses and
contributing value to the discussion. When you've
done that, people will be more responsive to what
you have to say are more likely to visit your site.
‣ Admit mistakes. If you overstepped the line in users
complain, apologise quickly. Never become
defensive and avoid confrontation.
‣ Don't spam. Never spam community. Don't visit
community, post an advertisement and never return
again. Don't post repetitively and indiscriminately.
Look for opportunities to add value to the community If you wish
you are participating in. The same is also true when to contribute
offering to write posts for editorial sites. to
Editorial sites can include news, reviews, and magazine communities
sites, as well as blogs. They tend to be topic orientated or editorial
(such as sports and) or lifestyle focused (like contents of sites it is
the over 50s). Some are run by large professional important to
organisations such as national newspapers, and others are add value to
run by enthusiastic amateurs, in the form of blogs. those sites.
Don't dismiss blogs. Some blogs have massive
audiences that exceed those of traditional websites. They
can be highly influential and have loyal subscribers who
take the bloggers opinion very seriously. These influential
blogs are also monitored by mainstream media, which
occasionally pick up on their stories.
Depending on the size and popularity of these sites it
can be hard to be featured. Many bloggers (especially
popular ones) are overwhelmed with requests to feature
various websites. If you wish to stand out from the crowd
you have to provide these bloggers with value rather than
just requests for free advertising.
In my personal experience, the best way of being
featured on an editorial website is to offer to write a
guest post. It is important that the guest post is not
Driving trafﬁc 147
primarily focus on whatever product or service you are
looking to promote. Instead it should provide value for
the bloggers audience. For example, if the website is
about health issues for the over 70s and you sell mobility
scooters, offer to write a post talking about things to
consider when buying a mobility scooter.
Most editorial sites feature a bio of guest writers that
includes links to your products or services.
Ideally the post wouldn't even directly mention your
website or products. That is because almost all guest
posts include a short biography about the author that
includes a link back to your site. There is no need to force
your site down the user's throat in the article itself. If you
do so the chances are your post will not be accepted.
If an editorial site that is particularly important to your
target audience is unwilling to accept guest posts, then
you might wish to consider becoming a regular
commentator. Commenting on other people's blog posts
is a great way of not only building your brand but also
drawing the blog owner's attention to you. I have found
that regular commenting on a blog can often lead to
Having a strategy for driving traffic to your website is
fundamental to your site success. It is also incredibly
important to ensure that the marketing methods you
choose generate the best return on investment possible.
That is why the following actions are important:
Action 1: Decide on budget and resources
Sit down as soon as possible and decide who is
responsible for the marketing of your website. Make
sure this is an official part of their job and that they
have time to do it. Also ensure that they have a budget
to do the job properly.
Action 2: Identify & target your audience
Make sure you have a clear understanding of who your
target audiences is and identify the various sites and
online communities they regularly participate in.
Action 3: Enable campaign tracking
Whether you use social media, paid advertising or
guest writing you need a mechanism by which you can
track the success or failure of these approaches. Ensure
you generate unique URLs for each campaign you run.
And so we come to the end of this book, but you have
only just begun the journey of ensuring that your website
is not a drain on resources but a source of income.
If you take away just a single thought it should be that in
order to maximise the performance of your website you
need to continually test and refine it.
So what are you waiting for? Stop reading and start
transforming your website into a money making machine.
About the author
Paul Boag is a website strategist and a director at
Headscape, a web design agency based in the south of
England. He started designing websites while working for
IBM back in 1994. This was in the days when designing
for the web was considerably easier. Since then he has
ridden the dot com bubble and co-founded his own web
Today he can be found advising clients on how to
better utilise the web, speaking around the world and
hosting the award winning boagworld.com web design
podcast. He also writes for his own blog and numerous
At the time of writing he is addicted to Twitter. He is
hoping the obsession will pass but he is beginning to
Thanks to Shutterstock.com for the usage of the imagery found
in this book.