>> Kirsten Wiley: So good afternoon and welcome. My name is Kirsten Wiley. And I'm
here to introduce and welcome Ric Merrifield, who is visiting us today as part of the
Microsoft Research Visiting Speakers Series. He's here to discuss his new book
"Rethink - A Business Manifesto for Cutting Costs and Boosting Innovation."
In "Rethink" he shows us how to rise above the clutter of your house to expose what
does and doesn't need attention in your organization. "Rethink" guides readers through
a strategic rethink to help set a course tailored for them.
Ric Merrifield is a former developer. He spent nearly 15 years in various consulting roles
helping organizations define and achieve their goals. Since joining Microsoft, Merrifield
has spent more than 10,000 hours as a business architect and has filed 12 patent
applications, all with the goal of helping companies rethink their operating models and
get out of the how trap described in his book.
He recently co-authored The Next Revolution in Productivity, a June 2008 Harvard
Business Review article focused on case studies that highlight needs of the organization
and the opportunity to rethink business operating models before making major
He's an alumnus of Lakeside School in Seattle and Georgetown University in
Washington DC. Please join me in welcoming Ric Merrifield to Microsoft.
>> Ric Merrifield: Thanks very much and thanks for coming today. The notion of me
being on the visitor series at Microsoft being also an employee strikes me as a little bit
ironic but at the same time being so focused on business productivity I often do feel like
a visitor in a lot of the meetings I attend.
I'm going to go walk through today some of the basic ideas of the "Rethink" book,
explain what the how trap is and talk a little bit about focussing on innovation, particularly
because today being at Microsoft Research, obviously innovation is a big part of what
you do as an organization. And one of my great frustrations in discussions about
innovation is people don't define it before they start talking about it. I'll offer thoughts on
how I define it. Whether you agree with it or not you'll at least have a construct with
which to have a disagreement with me on the definition.
So to get started, in terms of an agenda, I promise that for most of you, you'll see at least
one thing very differently today than you've seen before. And I usually have fun with that
particular example. Go through some actual results that organizations have realized
using this approach. And then I'll explain this core idea what I call the how trap where
people get so attached to how they do their day-to-day activities inside of work as well
as home it gets very difficult to contemplate doing the work differently.
A little bit more on innovation, as I said, getting into a definition. Give some examples of
innovation, to maybe spur some questions there. And then also get into why this notion
of the how trap is so easy for people to make a transition and get benefit very quickly.
And then I'll tie it into a lot of people are familiar with Jim Collin's book "Good to Great"
and the unifying principle there he calls the hedgehog concept. I'll explain how the ideas
in the book "Rethink" complement the notion of the hedgehog concept so you have an
idea how you can use hedgehog thinking complementary way with the rethink ideas,
then I have some closing thoughts.
Before I jump into changing the way you look at things today, I was talking earlier about
the process of getting a book published which was not what I expected. It was a lot
more work in areas where I wasn't expecting it. And at the very end of the publication
process, in about the March time frame the publisher, the Financial Times Press, came
back to me and said, okay, we've got this manuscript that's ten chapters long. We love
the idea of the how trap, and we love the examples, but chapter one is too long and
people will tail off and they're not going read the rest of the book because they'll lose
interest in the middle of chapter one. We said how do we solve that?
They said we think you need to break it into two different chapters. We said fine, we
think we can do it. We see two distinct ideas, break it into two different chapters. We
did that, spent a lot of extra time breaking the two chapters. I'm delighted how those
chapters ended up, making it an even better book than it was.
Then I went back to the publisher. I said this is at its heart presented as a business
productivity book. I honestly think going from ten chapters, adding another chapter,
there are 11 chapters and I think it's really bad luck to end a business productivity book
in chapter 11. Can't we find another way to have another section or some other way so
we don't end it in chapter 11. They said no, no. So we had some fun with that
conversation. But they rejected it. And it does now end in chapter 11. And now
particularly when I'm in the Pacific Northwest, most of chapter 11 is about Amazon.com.
So now I'm getting questions: Is it wishful thinking for Microsoft that Amazon is in
chapter 11. None of that either. But it's an 11-chapter book. We've had some good
feedback. It's a fun read. It's not very academic. The Harvard Business article I did
was much more academic and dense. And this book is designed to be a quick read.
And we've had some positive feedback from it.
So getting on to the way you see things today. I typically will ask for a show of hands,
and I'll spare you that today. But most people I know are not aware of the fact that
there's a very large arrow in the FedEx logo. And I find that when I tell people that,
about 50 percent of the people that haven't seen it before see it right away. If you don't
see it yet, there it is.
And so the reason for showing this is an example is that it's one of those things that
once you see it, you can't not see it. You're probably surprised you hadn't seen it before
if you hadn't seen it before today. The notion of the how trap I'm going to explain is very
similar for most people where the things that it exposes when you understand this
definition, the separation of what the outcome of the work is, what you're trying to do
from how you go about it, for most people it's a lot like that arrow in the FedEx logo. It
becomes so obvious that you see all these things all around you that sort of jump off the
page or jump off sort of the opportunity list in terms of, oh, wow, doesn't matter how we
do this. This is something we could do as the same as the next person in the cubicle
over or what have you.
So it's a fun example for a lot of people to take a very familiar logo and in many cases
expose something that people haven't noticed before. And if you hadn't seen it before,
you'll always see that arrow from now on. I promise in three to six months you'll
disassociate me from the arrow and just see the arrow as it is. And go from there.
So real results -- I do these slides both as a summary and then I'll echo them later after
I've explained this notion of a how trap. So I'll quickly give the three examples of cost
cutting, and I point out that we don't have logos of the company we achieve this with in
cost cutting, because typically the cost cutting results we get from customers are in the
neighborhood of 10 to 20 times what they expected. But they're so embarrassed that we
found them so quickly and that the magnitude was so much greater than they expected,
and partly also because some of them are public companies and they feel that there
may be some shareholder retribution, they don't let us name them.
So, unfortunately, the first example, very large manufacturer in Europe, you all would
recognize by name, they were trying to deploy a service oriented architecture which at
the time was the hot new thing. And the leader of information technology said, you
know, we think the best way to start this and really demonstrating business value to the
organization is by eliminating some duplication.
And we're pretty sure that in the area of verifying zip codes, a piece of software that a lot
of organizations subscribe to on an annual basis, costs between two and $3 million for,
where there's marketing mailers or for billing purposes, it's pretty important to have
correct addresses and zip codes or postal codes. They said we're pretty sure we're
investing in this thing twice. It turns out in three weeks we found significantly more than
that and realized in realized benefit in less than three months of over $40 million and we
redeployed 65 heads.
One of the important messages there was that the IT department got the money and
gave it back to the business and saying here's where there was waste and repetition,
now let's invest this money we've just given back to you in innovation and things that can
really help the company differentiate and have people a more positive experience with
the organization. So that was a great example. The next one's a big bank in the
northeastern part of the United States. Similar fact pattern where they said we've got
this thing called wire data services where we subscribe to stock quotes and investment
price instruments, Bloomberg, Dow Jones, Reuters, that kind of thing. They said we're
pretty sure we're wasting two to $3 million on that. It was sort of a humble expectations
project. In eight weeks we found over $23 million of repetitive spend, just because it had
been masked by, explained a little bit the how trap.
The last example I used I use for a number of reasons where zero savings were realized
a big logistics firm in the United States. They brought us in. They had five different
divisions of transportation. There was a truck division. There was a partial truckload
division. Vessels, airplane, and train were among the divisions they had. They said we
think there's about 10 or 15% overlap in the work that goes on in these different divisions
and we found in less than eight weeks it was over 80 percent overlap in terms of what
the work was, and there was an opportunity for significant consolidation and analysis of
best practices. They didn't realize any of those savings because it got hung up in office
politics. They knew they should do this. It was the right thing for the business but for a
variety of political reasons it got hung up and they didn't actually execute it and realize it
in the savings.
I include this here and in the book because one of my frustrations with some of these
business productivity books is everything is so rosy and all your problems are going to
be solved if you just read the book. Well, the ideas of the book are sound and have
been proven to be sound, but there's still things that can trip you up in realizing the
benefits. As a cautionary tale, there are things that will prevent you from necessarily
realizing the benefits as this particular organization did, much to our frustration.
When it comes to innovation, the customers where we help them out or work with them
love for us to use their logos because they're so proud of those particular innovations.
I'll spend time talking about both of these particular organizations more but ING Direct is
great not just for banking, but I think particularly in this economic climate, for a lot of
organizations, they found some really innovative ways to cut costs. And I often will point
out that people see cost cutting and innovation as mutually exclusive when in fact some
of your best opportunities for innovation are in the areas of cost cutting.
So for them as an operating model, as an on-line retail bank they decided, most retail
banks have their business as the branch-based business. We're not going to have any
branches in our retail bank. Eliminate a huge cost element from their operating model
allow them to deliver a lot more cost savings to their customers. They don't allow
customers to write paper checks. What that does is allow them to eliminate the
department that handles bounced checks which is a pretty significant expense for a lot of
The more important thing I like about ING Direct that happens at Eclipse Aviation, ING
Direct turns their backs on some customers. They say we have a really clear value
proposition. We deliver high interest rates in savings accounts and if that's what you
want, that's what we deliver. If you want us to hold your hand on the phone for customer
support, take your business elsewhere. If you want a branch, take your business
elsewhere. In having the ability to be that clear about what they're delivering and
connecting that with what a customer wants, I think particularly in an economic
downturn, is something to help our customers do more effectively to say this is our value
and this isn't our value.
If somebody wants that, they need to go somewhere else because this is what we're
doing. It may not be just one thing, I'll talk more about the notion of the hedgehog
concept, may be delivering multiple value propositions, but still be clear about what
you're delivering and not delivering.
Same thing at Eclipse Aviation, which is a company that was run by a guy named Vern
Rayburn, who used to be the original, one of the original marketing executives at
Microsoft back when Microsoft was based in Albuquerque. Rayburn realized there was
a huge opportunity for private jets to have -- so at that time the least expensive private
jet was about $5 million. And Rayburn realized that through a variety of innovations he
could come to market with a $1 million private jet and there was a huge underserved
market for people who were willing to pay more than commercial but couldn't afford the
$5 million private jet. And proved that there's a huge market for that.
I'm not going to spend as much time talking about some of the manufacturing-specific
innovations they did, but the point about Eclipse turning their back on some customers,
Eclipse doesn't allow you to customize the interior of your private jet. So a lot of the
costs in private jets are the kind of leather you have in your seats and the kind of fabric
you have and a lot of other factors go into the cost, extend the time line, and Rayburn
said you know what, I'm willing to bet there are people who just want private jets for
transportation. And even though they want it to be nice inside, they don't care what
leather it is. In fact, that was one of the biggest areas of cost cutting for him in the
overall processes. This is what we're delivering. We're delivering a low cost private jet,
if you need customization, call the next, you have to spend $4 million more at least to get
The other one that he did that's a little bit more disruptive that wrinkles some people's
noses, eliminated the bathroom from the jet. These are short haul flights, not go in the
air more than two, three hours. People need to plan ahead. It eliminated huge amount
of weight from the jet, helps with fuel economy and delivered a lot of other things in the
process. But getting people, people assume you have to have an bathroom on an
airplane and Rayburn proved that was not the case.
Getting this core notion of what I call the how trap, very much a human condition, affects
all of us. I'll give you some examples here. Some of the outside of the workplace.
Some more in the workplace. Most of us have a favorite restaurant or favorite
destination. As time marches on, we associate the route we take with the destination, so
much so that when we get in somebody's car there's this why are we going this way
conversation that most of us had with somebody at one point or another. Really doesn't
matter how you get there, as long as you achieve the outcome of the destination on
Similarly, the Fosbury Flop, a guy named Dick Fosbury, an Oregon based athlete in
the '60s, was a high jumper. He realized in the '60s, most people at that time did the
high jump with the scissor kick, much like people do the hurdles still today.
He said if the outcome, the measurement of success is getting higher over the bar than
anybody else, he figured out that turning 180 degrees the other way and going
backwards over the bar got a better result and got him an Olympic Gold Medal. And the
Fosbury Flop is now the way people jump over the high bar. Most people don't
remember it used to be done any other way.
As an individual accomplishment, where one person tested does it really matter how I
get over the bar, no, I could do better doing it another way, Fosbury is a great example
of an individual getting out of a how trap.
So those are non-work examples that may resonate with you. Now more into the work
environment. If I am doing a business requirements document, or if I'm a consultant
doing something with a company and I don't know anything about the company or even
the department that I'm talking to, if I walk up to a person at the fax machine and say,
look, I'm just trying to get a sense what your work is, I may be trying to help optimize it, I
may be trying to cut costs or innovate, tell me what you're doing, and the person at the
fax machine is going to say, probably tilt their head sideways a little bit and say, tilt their
head sideways and well, I'm sending a fax.
A lot of business requirements gathering discussions going on today, the person is going
to say, okay, is sending a fax a necessary step in your job? Do you have to do that to be
successful, or is it one of a sequence of process work flow steps. And in that
conversation the person is going to say, yes, it's a necessary step for me to do my job
successfully. If I were to say, okay, what you're doing here is communicating the status
of something or maybe confirming the order, how you're accomplishing it is with a fax
machine by disentangling, in a very simplistic way disentangling what I'm doing from how
I'm doing it, I can now test with the end user, if you're communicating status, if what
you're doing is communicating status, does it really matter whether you use a fax
machine or whether somebody else does it or whether it's outsourcing; if you accomplish
that outcome with whatever performance objectives metrics you have, can we change
how it's done? Maybe we even automate it.
Typically that covers about 90 percent of the time I found that conversation ends up
convincing the end user that it doesn't matter how it happens. It's fine that they keep
doing it that way, but if there's a different way that helps them be more successful or
frees up time, they're fine with it. And it's not -- if you were to challenge -- if the fax
discussion, if the previous method was, okay, it's a requirement to use a fax, well, I think
we can automate it. That's a much more sort of collision-oriented conversation where
I'm the non-expert challenging the work that the person's doing. Like how dare you
suggest that I'm the expert at my job, whereas when you separate this what and how,
they're like, okay, now I'm looking at my work differently. I realize that the fax is the how.
And it may not matter how I do it. It's a much more elegant way to test whether it
matters at a tactical level how the person's doing their job.
The flip side of that example, if United Airlines were to call Microsoft today and say, look,
we've got this, we're an airline. We have this -- people have to check in for their flights
and it's a very commodity function for our industry, we think we can cut some costs out
of it, can you help us?
A lot of the Microsoft people and partners who get that call today will say, well, Microsoft
doesn't have a product called Flight Check-in, so I don't really know what to do with that,
can you give us some more information. And in some cases the conversation will stop
If United Airlines said, okay, we have this thing called flight check-in but it's made up of
three separate steps that don't have anything to do with the airline industry. There's
confirming a reservation, conducting a survey. And if they've got luggage there's
If the problem statement is presented to Microsoft or anybody that way, okay, we've got
lots of case studies on reservation surveys and logistics, now let's get specific about
what specific performance improvements you're looking at; is there a customer service
change; can we talk about your technical architecture and get into a much richer
conversation just because the irrelevant attributes of the problem have been stripped
away and you can really hold it up and say it's reservation surveys and logistics and
that's the problem I'm trying to solve and then get specific about metrics, performance
change, customer service, all the things we can help them with that we have mountains
of case studies from best practices to technical architecture, to service oriented
architecture, to cloud things.
We now know for that particular example it literally doesn't matter who does the work. It
could be us in our hotel room. It could still be the airline employee behind the counter.
Doesn't matter where it happens or it doesn't matter what the technology is, as long as
the outcome of those three things is getting accomplished.
So those are some orienting examples on this basic notion of the how trap that we found
to be very simple but effective in terms of changing the conversation, making sure
people are really focused on the outcome of their work. So as a little more detailed
example, I find that once you sort of can spot these trap verbs, and I've got some of
them listed here on the left, it's easier to see them and say that's a how verb. What is
the what verb that I should be using in place of that, that it's fine to know that that's how
it's happening today, but I should have a what verb that connects to the how. This is an
actual process work flow diagram from a very large local company that sells insurance
for the products that it makes.
So I circled -- so I've got mailing the agreement here, faxing the agreement. So they're
all how verbs. And over on the side I put communicate, confirm, register, complete and
transport. The automate maps to the none. Automate is a funny word. It's not really a
how word or what word. So when I talk about that, some people have tried to push
automation in one category or another. And it's really neither, which is why I call it a
The last point here, when you sit down with somebody to talk to them about their job,
they're going to be inclined to use how words because they're so used to describing how
they do their job on a day-to-day basis. It's not really their job in most cases to describe
their job to somebody else. So don't fault the person for using how language, because
it's not really their job to do that. And the better able you are to help them separate the
whats from the hows, the quicker more rapidly you can get to more objective
conversation on where they are.
A quick note. This is not the same as process. And some people confuse process and
this what language that makes up the heart of what we call business architecture.
Process is work flow. Really the animation of what the work is.
And so in this case you see a start and an end arrow. That's a great process work flow.
One of the issues with process is because it tends to be encumbered with this how
language, it tends to be very subjective. So if you have five different people in the
department that are creating the insurance quote, if you sit down with them individually
and have them map out a process work flow, you're probably going to get five different
pictures. Which, if you have the definition of what the work is first, can be really helpful
to get those five different perspectives and say one person may have more air handling,
one person have more interaction with the partner and understand what some of those
different permutations are. So when you're looking at, one, what's most valuable, two,
how things are performing, you have some different inputs and some of the different
perspectives from the different departments.
What tends to be a far more durable lens than the how language. So, for instance,
sending a fax to the insurance agent, happen to know this company has been selling
insurance policies for over 80 years. If that is communicating the status of the insurance
policy to the agent, they probably have been doing that communication for 80 years.
They haven't been using the fax machine for 80 years. And so seeing that you can have
this more durable lens on what the work is and the more transitional
implementation-based how language on top of it, makes for a really elegant connection
between process and business architecture or the whats.
Why this is effective. Now I'm going to take this great quote example and build a map of
the whats and then show an example of one of the heat maps that we'll do that really
brings to life some of the discussion that we have with organizations.
Because you're having this what/how conversation, it's a very different conversation from
what the end user is used to having. So they often get immediate value from it rather
than just, oh, another process discussion. I just had Accenture in here six months ago
and they don't see it as value-added activity. The deliverable they get from this in half
an hour to 45 minutes typically is a different lens on their work and they're usually
excited to see it and more eager to participate in it.
The other piece like the fax example, you often don't have to be an expert consultant to
go in and sit down with that person to have the conversation. Just being able to spot the
how verbs and test, figure out what the what is enables people who are not business
consulting experts, even the people in the department already, to flush out some of the
differences in their work and really start to test whether it matters how they do certain
things, which then informs or helps inform the prioritization of what work is most
important for them, what projects are most important.
And then once you've teased out this what, which is an important first step, then you can
start to attach things like business value, performance, maturity of the work,
interconnectedness, which I get further into in the definition of the book, but attaching
those things can really help again inform prioritization of work.
So now I'm going to overlay on top of the hows, the whats. It's not important to
everybody that we read those. But I overlay those. In the middle there's a rekey into an
application. So I'm not going to move that one over. So there's six child whats that
make up the create insurance quote. I'll pull it off to the right so you've got the six
children inside of a parent. That by itself is not particularly magical to the end user at
this point in the conversation. Okay, those are whats. That makes sense. I get the
difference between what and how. The more pivotal point in the conversation is when
you ask them what's most valuable, what's least valuable, how are things performing,
what's performing well, what's not performing well? It turns out that most end users
have a pretty good sense, whether it's assertion-based or metric-based, have a pretty
good handle on what's most valuable, least valuable, what's performing well and what's
not performing well.
So if you add coloring -- so in the case of performance, each shape, each child here, so
the six children have an outline color that denotes performance. This is a green
performance which means it's performing well. A pink border means it's performing
poorly. The parent create quote has the same color scheme. So a pink border means
it's not performing well. The interior color, the fill value denotes responses to business
value. So if somebody's got a pink or red interior, it means it's high value. If it's got a
green fill, it means it's low value. So once we've got this view, someone says, okay, I
think we should make a big investment in the area of creating insurance quotes, I'm not
convinced just on this alone -- if the performance is bad, that's one thing, but if it's only
medium value, there may be other areas in my organization that I want to look at.
So I probably need to get a larger organizational view to say if it makes sense for me at
all to even invest in creating an insurance quote. Then if the decision is, yes, we
probably should invest in creating an insurance quote, then what? Now we've got -- we
see a couple of different combinations of children. There's one in the top right that's got
a pink interior and a red border which means high valued underperforming. It's tempting
to look at that and say, well, that may be the one that will pivot or cause the parent to
perform better since it's the most valuable and underperforming of all the children.
If I also am thinking about from a cost cutting perspective, maybe I want to look at the
most low value thing that's already performing well and say maybe I could outsource
that. Maybe there's a partner, if I'm looking at, say, sending the agreement, maybe
there's a third-party fulfillment house that can do that low value activity better, faster,
cheaper than I can. If it's already performing well, then why not. The next step beyond
the innovation, sort of beyond innovation, where would I improve it, testing, what children
would cause the parent to perform better. That's a conversation that most people
haven't, don't have a map of that's handy for them today to talk about. They're like, oh,
yeah, I understand the notion of causality and parent/child relationships. If I can test
which child will cause the parent to perform better, which one I can go ahead change
because it doesn't matter how it gets done, I can outsource it, a very different
conversation from what most end users are used to having.
Then as the next step, if they don't already have concrete metrics, attach metrics to it
and say, okay, if I know what specific performance measures I think I should get from
this, what's that going to cost. Is that a people change, a process change, a technology
change, and get into a cost benefit analysis of the situation. And then broad brush
strokes, the child metrics and coloring should align with the parent.
So with the parent that's medium value, it makes sense to have an array of value and
high value and low value work, similarly with performance, with a pink border for creating
a quote, it makes sense to have a blend of high performance and low performance
things. That's something that you tend to have to eyeball, because sometimes there's
going to be the five of the children will only happen infrequently and one of the six
children will be the one that really gets done all the time. That will be the pivot always
for the parent. And that becomes sort of a matter of familiarity with the work. But, again,
most times the end user knows which will cause that performance pivot.
Hopefully now you're starting to see the FedEx arrow analogy and you're getting the
difference between as simple as it is, the pulling the what away from the how makes for
a much more objective transparent conversation with people about not only what the
work is but then where do you need to prioritize to make changes.
As I said, I did some innovation-specific slides for this crowd. In my definition of
innovation, or when I hear people talk about innovation, there's a lot of people who I
think are talking more about creativity, like lightning striking, like we've got this great
idea, now what are we going to do with it.
I think of innovation more in terms of being on the change spectrum. So when
somebody comes in and says we need to cut costs or we need to respond to this
competitive pressure, my response is: Okay, do we need to do like a minor tuning
scorecard change is that the right kind of change to respond to the challenge? Or is it
more like a big project, like a remodel, where the work's going to resemble what was
there before but it's going to be clearly different. Innovation is on the far end of that
spectrum where the outcome is the same, and airport check-in is not a bad example, but
it doesn't resemble what was done before.
Checking in for a flight on the Web doesn't resemble the experience of talking to the
airline employee but accomplishes the same outcomes of confirming reservations and
conducting a survey, managing logics, if you've got that.
Asking the question, is the right response to do the minor change or innovation help
people get innovation on the table of every conversation. So it's not a specialty activity,
it's an every day when I look at a problem is innovation the right answer. And then the
cost benefit equation you could see, so my examples here we've got simple change, free
checking, is something that didn't take retail banking that much of an effort to do.
And it probably did improve their numbers. On a project basis, automated bank
statements, that's probably something that took a fair amount of work for banks to do,
and then innovation, the ATM where you actually take the bank teller and push the teller
to the outside of the bank or even to your 711 or what have you, you're still getting your
cash but the experience doesn't resemble going into the bank and talking to the teller
and the teller knows who you are and maybe knows your dog's name, radically different
That's an admittedly a simplistic definition but that's how I define innovation. From there
I think of innovation in a couple of different categories. You've got just simply a better
how it gets done versus there are new, actually new whats people are doing.
So in the better how, Netflix is simply a better how I rent videos or any other service like
that. You don't have to go to the store anymore. I get my movies through the mail. I
send them back whenever I want. The additional benefit of not having to pay late fees is
just a perk.
ATM is a better how people can get money because. They can do it anywhere they are,
don't have to be attached to the bank hours anymore. The nano car from Tata Motors
they're selling for $2,000, just a cheaper car. They figured out a way to slash the cost
out of a car and both from a manufacturing perspective as well as a price point it's just a
better method for a lot of people. May not be as safe as it needs to be yet, but it's a
FedEx and Amazon sort of straddled the better how and the new what and I'm going to
come back to them. The new whats -- I actually think e-mail is in many respects is a
new what in the sense that when we first started using it we had this mentality of -- and
I've got this graphic on the top from the New York Times, it's brand new but make it
sound familiar, a lot of innovations like the automobile, when it first were introduced,
people didn't have any -- why would I buy this was the initial reaction.
And until people called it the horseless carriage people said, oh, this is an alternative to
the carriage. And then, of course, cars evolved to be much more than that.
I think with e-mail, people started, the metaphor was, okay, paper-based letters and
memoranda, the notion of how we use e-mail today as a replacement for dialogue and
almost replacing the phone in a lot of cases in terms of quick questions, it's evolved in
really a new way of how we communicate with people, sometimes work people,
sometimes socially, and in many respects it's a new what.
Fan base is an evolution of social networking where they take, it's a Wikipedia for
nonprofessional sports essentially. So high school and college sports, it gets people to
register their high school soccer team and have all the names of that, and allows a
connection to people who have some association with each other that was never really
possible as rich an experience as you get from fan base as they've done.
OnStar, all of the onboard information in cars where you cannot only have a lot of
information pushed to the driver about the health and wellness of the vehicle as well as
finding the nearest McDonald's, whatever they need to do, that's something that people
just I don't think could have really even imagined 20 years ago.
The other piece with OnStar that I think comes out in January is that when there's a
collision with OnStar, they collect some crazy number, like 100 pieces of information
from the vehicle, and immediately transmit it to the OnStar database, which pushes in
some huge repository of crash histories, which then crunches those numbers and then
immediately or as quickly as it takes to crunch it, they push it out to the emergency
service organizations so they can tell the fire department and the emergency response
units, say this is what kind of crash it is, based on our data, this is the likelihood of
serious injuries. This is the kind of equipment you should bring. This is where it is.
You'll need scuba gear, all that kind of stuff, they'll do it an instant sent out to emergency
response teams within seconds of when the accident actually happens. These are the
kinds of things that OnStar's doing is exciting and they're actually using Microsoft Earth
for the images.
Coming back to FedEx and Amazon, I think both of them are examples where they
started with something familiar. Amazon was selling books. FedEx was saying
overnight mail. But both of them have evolved into something so much more, we really,
we don't really think of FedEx and that type of service as an alternative to mail. It's just a
different way of -- it's a different service.
Just as Amazon has evolved really into more of a software company than a book, at
least in their description, they're more of a software company now and the backbone of a
lot of their interactions more than just a bookseller. I don't think FedEx or Amazon would
have gotten there if they hadn't started out with something familiar and built on that.
Operating models. Innovative operating model I spent a lot of time with the Newman's
Own organization, a number of years ago. And what was interesting to me about them,
and I'd read about them in the New York Times, a lot of people from the outside world
think of them as just another food service company that happens to give all their profits
to charity, which is again broad brush strokes that's true, that's what Newman's Own's
But the more interesting thing to me was that this company with revenues of over
$200 million has fewer than 20 employees. For them, they didn't care about running a
business. Paul Newman only cared about having really high quality recipes. They have
a super kitchen in their office and giving profits to charity.
Everything else for the company was outsourced. So a lot of times when I go into
customers I'll say Paul Newman was really clear what he valued and didn't value in
terms of running a business. Are you that specific? Do you just keep growing your
business because you think other companies in your industry are doing that, or are you
being about specific what's most valuable to you and your customers. In that sense it's
a good example.
The Costco example that I like to use, if most people here obviously familiar with Costco,
if you compare them to Sam's club, what goes on in both of those organizations is nearly
identical. But their operating models and how they get their profit turns out to be
radically different. Sam's club is a very traditional retail model where they care about
whose products are on their shelves. They negotiate really onerous supply agreements,
drive the prices down in terms of what they are willing to pay and they put margin on top
of that. That's where Sam's Club gets their profits.
Costco, by contrast, doesn't make any money selling their products, by choice. They
also don't care whose products are on their shelves. They'll put anybody's products on
their shelves. One week it may be Hellman's Mayonnaise, may be Best Foods the next.
They make their profits from their membership dues. They light their stores brightly,
friendly employees, giving food away all the time. But if a high priced consultant was to
go into Costco, drop in retail industry best practices, they would destroy value by not
knowing the differences in the operating model of Costco.
The other example I like to use that I get some nose wrinkles on campus when I use this
example. If you think about the Coca Cola company, they're a company that doesn't
make anything. They create intellectual property. Everything they sell, including Desani
water, starts as a powder. I like to think of it you just add water water. They sell it to
bottlers and distributors. They then put it into the containers. They sell it to the end
They create intellectual property. Spend a lot of money on branding and marketing. I
often ask groups, customers, name another company that you can think of that doesn't
make anything, doesn't really sell much to their end consumers, and only spends money
creating intellectual property and branding and marketing.
Microsoft. Right. So the good news is by seeing the improbable comparison of
companies that aren't in the same industry, there are all kinds of opportunities for best
practice analysis between Coca Cola and Microsoft, and we have had some of those
conversations by stripping again away some of this how language and some of the
industry expectations, saying these are companies in completely different industries that
are actually doing very similar things in many respects.
All right. One more slide on operating model. This is -- I can't take credit for building this
slide. The woman who wrote Blue Ocean Strategy, Renee Moborn, did this. But it's a
good example in my opinion of looking at Starbucks saying they listed out a bunch of the
qualities that go into the experience of buying a cup of coffee, quality, speed,
convenience, fashion, selection, ambiance, price and location and service expertise.
They said, okay, what does the rest of the industry put emphasis on and what are we
going to emphasize differently? And they picked out different things.
Here's what the industry cares about. Here's what we care about and here's where
they're different. One of the ones that often comes up as a question mark, in the green
circles are Starbucks and the red squares are others like Dunkin' Donuts and Denny's,
other places where you can get a cup of coffee. They'll say: Why is price down on the
bottom for Starbucks and Denny's is high up? And the response is because Starbucks
doesn't care how much they charge for their coffee. Denny's cares a lot. So that's
the emphasis point. It's not that Starbucks is cheap. They don't care how much they
pay for their product. It's not the value proposition to their customer. Each of those
decisions is what they emphasize what they value, versus what they don't care about.
That turns out to be a picture, a fairly illustrative familiar example at al little more detail of
what I've been talking about about what an organization cares about and what it doesn't
Some additional trapping ideas I like to talk about. I mention the flight check-in that vets,
there's no need to go into that as being an airline-specific problem to see where the
opportunities for improvement are. Value confusion, I'll go back to Starbucks for a
moment where you talk about Internet connectivity in a coffee shop.
The question I often will ask people is do you think that's more of a marketing expense
for Starbucks where they're trying to bring people into the store or keep them there so
they buy more coffee and donuts and CDs and newspapers, or is it an actual product
itself if they're charging for it where it's a source of revenue and that's a stand-alone
metric by which it should be judged and I'll often get a lot of feedback from people, it's a
marketing thing, no it's a sales thing.
My response is it doesn't matter. What matters is that it has a specific job and it gets
measured for what its job is. So when times get tough, you say let's look at this WIFI
thing in the store and say if its job is to bring people in sell more coffee donuts and CDs,
is it doing that job and if it's not how do we treat it differently? Should we eliminate it or
we should invest more heavily to get that metric up. Or if it's a product, is it driving
enough revenue for us to keep it as a product. Is it profitable enough to keep it as a
product? In reality, it's a little bit -- I know Starbucks is a little bit of a blend of both when
they're still charging for it. That's one of the things I see a lot of work in organizations
where people haven't said yes we have to do it and they'll say, okay, what's its job. Well
you know its job. No, I don't. Tell me. Is it a marketing thing, a sales thing? Is it a
commodity thing? And be specific. Have a specific metric about what its job is.
Last the one, Toshiba had in their consumer laptop repair group several years ago a
problem where it was taking several days, actually over nine days, like 11 days, to turn
around a consumer laptop for repair.
And then and now most people can't be away from their laptops for 11 days. So they
looked at it. They said what's going on, why is it taking 11 days for us to turn this thing
around. Long story short, they figured it was taking only 45 minutes for actual laptop
technician repair time to fix the laptop, and all the rest of the time was the orchestration
of getting the laptop from the consumer, getting it to the repair station. If they had to
order parts, they'd have another UPS truck go get a part and bring it back and the
laptop, the technician could fix it, and they'd call UPS again and have them deliver it to
the consumer. That's what was consuming all the time. They said hang on this is not
really a laptop repair problem logistics component. It's really a logistics problem that has
a tiny laptop repair component.
We're a technology company. We're not a logistics company. We should outsource this
whole problem to a logistics provider, which they did. UPS took the whole thing on,
including hiring the laptop repair technicians and ended up getting average laptop repair
time down to a less and a day and half because UPS is as good at logistics. A lot of the
problems have the wrong labels on them. That's a good example of one where it had
completely the wrong label. They were chasing after it as a technology problem when it
The why now? Some of this is fairly obvious. For the last 20 years a lot of industries
have been in constant growth mode. Constantly just keeping up with the growth, moving
as fast as we can. Waste and inefficiency is creeping into my operating model. That's
fine, we'll get to it eventually. We're there now.
What most people I'm finding are saying, well, we've got some waste and inefficiency
that's crept in. Turns out the waste is gigantic. That over 20 years most people expect
X when they should actually expect to find 10 X waste and redundancy, inefficiency
processes, even drifting the business into an operating model that has no direct
connection in terms of what their customers value, not really knowing who their
customers are and are not. So it makes it very difficult for them to tune where they need
to cut costs, let alone where they need to innovate.
A little bit like the FedEx era. Once you see it, it becomes so obvious, it's a little bit
tempting for organizations to blame people and say how did you not see this? You're
stupid. That's nothing good will come of that conversation. Because it is such a different
lens for most people, because things become so obvious for most people, the message
is get on with it. Realize the savings. See the opportunity for innovation and take
advantage of it. The final point is that because you'll see so much repetition, most
organizations will see so much repetition work, it's a great -- not just for technology, but
also best practices.
I mentioned this notion of the hedgehog concept from good to great. A lot of people are
familiar with it. But if you're not, the hedgehog, the unifying idea of this particular book,
Jim Collins says you need to focus on this intersection of what you can be best in the
world at, what you're most passionate about, what drives your economic engine.
In the center of that is what he calls the hedgehog. That's the singular thing you should
be focused on as an organization. I think it's good advice to help organizations be really
clear about that particular, what really should we as an organization be focused on.
The reality is most large organizations are not that simple. This VIN diagram has seven
different points in it. The hedgehog is just one point. It's possible that what drives your
economic engine can intersect with what you can be best in the world at but is not
something you're deeply passionate about.
That sort of stuff happens all the time. Concur Technologies is a local software
company that ran into it a number of years ago, where their licensed software business
was in that part of the VIN diagram, they wanted hosted expense management software
to be their hedgehog. They for a number of years had to make trade offs, we have to
keep funding this one thing that isn't our hedgehog because it's still driving our economic
engine. But we also still have to keep investing in our hedgehog. And I'll tell you that
you're going to find things that touch all seven areas of that VIN diagram.
And one of the things that you'll find hopefully helpful from the "Rethink" book by doing
the kinds of heat maps that are discussed in the book, you can have this conversation
about parent/child relationships, value performance maturity and say whether or not it's
in my hedgehog, is it something that I need to be investing in, or is it something I can
ignore, just because this established account revenue gaps is high value and
underperforming, if its parent is performing well, I'm not sure I should invest in it. Maybe
I can ignore that thing and have those kinds of conversations as a complement to the
So that's a simple overview in terms of how to think of the heat mapping and
prioritization of work as a complement to Collins' hedgehog.
All right. Some closing thoughts, and we can open up to questions. It's tempting for a
lot of organizations to chase after the really high value work. The problem with the risk
with that is that if it's high value, if you make mistakes, the mistakes can really be painful.
My recommendation is start with the lower value areas of the business, because if you
make -- as you learn any new methodology you'll have some lessons learned. Learn the
lessons in a low value area, that's going to be less impactful, get familiar with what it
does as a method that may be different from six sigma or LEAN or process
reengineering or whatever method or software development lifecycle you're using. And
then once you've got that, then point the lens at the higher value stuff. And I think you'll
have better success with that.
A lot of managers are going to be familiar with the notion that somebody runs in their
office and this is the number one most important thing we need to be doing in this
organization; if we just fix this, everything will be fine. A lot of managers don't have a
vehicle to push back and say I'm not sure that's right; are you sure that is going to cause
my department to perform better? That's the one thing that's more important than
Heat maps prove to be a way for the person to say, okay, show me a heat map that says
not only this is the number one or these are the two or three most important things, but
show me in parent/child relationships why the reason for your assertion that this is going
to flip the performance of the parent or the department or the organization as a whole,
much more objective, less confrontational way to have that conversation.
And if it turns out that the person is wrong, that their one thing isn't the most important,
instead of just saying you're wrong, they'll go, oh, now I see why my shiny object project
is maybe not going to be done until next year. Here I now understand why you've made
the decision to invest in these other things. We'll eventually get there, but it's actually
not the number one most important. They're not just getting dismissed, they're being
heard, and they're also seeing why prioritization decisions are being made in the
organization. It's a lot like the FedEx era. Once you see it, you can't not see it. And it's
tempting to judge people for not seeing it. No good will come of that. Just get on with it.
Expect people to see things they've never seen before. And have fun with it. It turns out
to be fun. And getting people talking about the how trap, applying it to their work, does
tend to be fun.
And hopefully this has been an interesting talk for and you can tell me some of your
stories about getting out of the how trap, you can find me on campus, but there's a book
website rethinkbook.com or send it to rethink Microsoft.com, if you don't want to use my
name in the e-mail for some reason. So with that, hopefully this was interesting, a good
use of your time.
If you have any questions I'd love to hear them.
>>: I have a good example of where almost happened. It was pretty close. I was
actually meeting with British Airways earlier this year. They mentioned they were in final
discussions to buy a telepresence company.
>> Ric Merrifield: Buy a what?
>>: Telepresence company, video conferencing. What we've been saying, I'm in the
travel group here, what we've been saying is we don't want to fly anywhere. We don't
care about flying. We don't want to be in hotels we don't care about any of that. We
care about getting business done. Those are the most likely companies to step in. And
it's just interesting that they were really thinking that they're not in the business of flying
people, which I think is a pretty big change.
>> Ric Merrifield: Right. Yeah, that's what I call paving the cow path where people are
so attached to some outdated notion they're trying to optimize something that actually
isn't -- isn't a valid assumption anymore, necessarily.
I think you're right. Travel is something that people just got so used to -- yep, I've got to
fly to go to solve that problem, rather than we can do it on a con call or video
teleconference. I think it's going to be one of those 10-year overnight success stories
where people in 10 or 15 years are going to be so much more accustomed to having
teleconferences for things that were just assumed that were necessary as a face-to-face
>>: I think we're 10 years into it.
>> Ric Merrifield: Fair enough. Twenty-year overnight success. Other questions?
>>: You talked about cutting costs.
>> Ric Merrifield: Cutting costs and innovating.
>>: Do you talk about laying off employees?
>> Ric Merrifield: The question is whether it talks about laying off employees. No,
there's not a specific topic about laying off employees. But in this economy, that
obviously has to come up.
And there's a slide that I often will show that I didn't today. One mistake I find a lot of
organizations making right now is all the doom and gloom in the news, the recession is
affecting everybody equally. Turns out that about 20 percent of industries have grown in
the last year and a half.
So some companies that sort of get into this cost cutting mentality is the wrong direction
for them to take. They're not even realizing necessarily that their business is growing
and doing quite well.
So my message is, sure, if you're an automobile manufacturer in the U.S., it's probably
going to be a pretty bumpy road for you for a long time. And there's nothing I can do to
help that. And I'm unfortunately going to blame them for a lot of the mess they're in. But
if it ends up having to be a message of laying people off, if that's the right thing for the
business, the organization has to do the best thing for its shareholders, and there's no
easy way to ease people into that message.
>>: So it seems to me the most obvious example right now to come to mind about the
trap like this is journalism, great journalism, newspapers. If you were to get hired by the
Seattle Times, please, please, please, and go in and you say let's think about what
you're doing, what your delivery is, what would your advice be in terms of their strategy
>> Ric Merrifield: That's a good question. Did everybody hear the question? Okay. I've
actually written a couple blogs about this, because I think there are a variety of industries
that are caught up in the trajectory of how people are consuming information, but also it
does interweave social networking, so where are people going to be -- it even slices up
by demographic. So I'm a 43-year-old guy who is probably never going to never not
have a subscription to New York Times. I always want to touch the paper. That's my
Also, pretty good bet that my eight-and-a-half-year-old son is never going to have a
subscription to a newspaper. So understanding, getting the Seattle Times, the New
York Times, Facebook, LinkedIn, to say by demographic how are they consuming
content today, how has it changed in the last few years, because articles are getting
smaller and smaller. People aren't wanting to read the full-page New York Times
articles as much as we used to, what is the attention span?
How much do people want to read by content? Do they want to read the top 10 stories
like what's most urgent? Or are they like me the person who will see an article off to the
side of the page I never would have searched for that. There was an article I remember
in the New York Times about baby duckbill platypuses, which are called puggles, I
learned, that the duckbill platypus is the perfect intersection between mammals and
reptiles. It's a fascinating study on how much scientists learned. I never would have
searched on that. So the question of how do you make information available to people
that's interesting that's not on the top 10 list. I know that's probably a longer winded
answer to your question.
>>: You're focused on your audience.
>> Ric Merrifield: Know your audience. Look at the trend of your audience, how they're
consuming information differently, how their social networking patterns affect, like these
people sending articles to their friends and family.
All of that, I think, is sort of that's the 10 and 20-year look at information and content
consumption, is to say it's not so much about the newspapers today and the magazines
today and the Facebook today, it's like what's the trend and the devices people are
going to use to consume that, is it going to be like text messaging? Do I want to get a
text every time the Yankees win a baseball game? I don't.
So having those different options available. Even some of the lessons that I think we're
learning from online banking where the customer gets more control over what gets
pushed to them, where they want help, where they don't want help. I think that's also
going to be a factor in content. I think it's a giant problem that's just starting to take
shape. And it's not the newspaper industry. It's a variety of content, social networking,
timing and demographics that will inform where we end up 10 or 20 years from now.
And we'll all be surprised. That's a longer answer than you probably wanted.
>>: Have you done studies specific to Microsoft versus our competitors in terms of
innovation and cost cutting?
>> Ric Merrifield: Have I done research inside of Microsoft specific to cost cutting
innovation? I've done a lot of work inside of Microsoft in terms of where, helping
divisions make prioritization decisions in terms of where investments make sense, both
from a cost cutting perspective and an innovation perspective.
I have not specifically had the competitive discussion. But I know -- so I come in as the
non-expert, which is sort of ironic, but I help expose this lens, this view of their work
they've not seen before and help them talk through the child/parent relationships, the
dependency discussions, because they've never seen these heat maps before. Wow,
I've never had this view of my work. Help me make decisions about how I prioritize. I
know they've had competitive discussions and had more cost cutting innovation
discussions. I'm more of here's how you read the map and here's how you should think
about prioritizing it, both high, low value stuff. Is it a people change? Process change?
Technology change? Have you already got projects funded in this area? Should you be
rethinking your FY10 budgets for this area now that you've got this lens. That's the kind
of conversation I have more internally. Question in the back.
>>: I've been thinking, if you have everything to lose or nothing to lose, rethinking is
easier. If you watch children play, they always -- they will do things an adult will never
dream of doing. And if you're in a company, in a bankruptcy --
>> Ric Merrifield: Like wearing a Halloween costume to school on a non-Halloween day
I'll go as a space man today. That's something you or I would not do, wear that to work.
>>: Exactly. Being in a bankruptcy or impending bankruptcy like this country has seen,
and then all kinds of radical decisions are okay. They're forced upon.
>> Ric Merrifield: That's right.
>>: So I'm looking around and I'm going there's that when you have nothing to lose and
you have everything to lose. And then you have situations like ours, the company, we're
comfortable, we're surfing okay, and there's enough cash in the bank and there isn't
much to lose.
>> Ric Merrifield: I wouldn't go that far, but...
>>: Rethinking becomes difficult.
>> Ric Merrifield: It does.
>>: In that kind of environment.
>>: So the urgency is missing.
>>: Yes. People [indiscernible] with rethinking is what I'm asking.
>> Ric Merrifield: The analogy I use, parents who have kids that are probably 10 or 12
years younger will recognize in the movie "Finding Nemo," towards the end of the movie,
Nemo has gotten out of his predicament, swimming home with his dad and he ends up
getting caught up in a big commercial fishing net.
And the fishing net's being lifted up, and they're being successful in catching all these
fish. One of the lessons that Nemo had learned on his adventure was if he told all the
other fish to swim down in the net they would be able to snap out of the net. He tells
them all to swim down, they swim down, they break out of the net and all the fish are
I think what happens in a lot of large organizations is you have people, it's a little bit like
the shiny object project that I mentioned earlier, they think they're doing what's best for
the company but what they're doing is going one direction, what somebody else is doing
going in another direction. So there's a lack of unity in terms of the direction of the
efforts. And a company of our size, you're going to have to have multiple efforts going a
variety of directions.
But I see a lot of different efforts that don't really align and snap to themselves. And they
end up in some ways conflicting. I think that's the real opportunity is to get that visibility
to say which projects really need to be aligned, which ones are actually pointing at
So Robbie Box stuff at XBox is going to be talking about something that's very different
from what Chris Capp is doing with Office 14 Wave 14, but they should be as alined as
they can in Wave 14 in terms of the value proposition, the demographics, the message,
how we take that message to market.
And I'm using those as random examples. I'm not saying Robbie or Chris are doing
anything wrong, specifically, I'm just saying that I think there's an opportunity for much
greater alignment of effort so we can be more unified in going after what we're chasing
But you're right, nothing to lose/everything to gain is much easier to do than rethinking
than going into something that's already very complicated, probably very political and
testing some of those things.
I do find that the heat maps tend to drive a more objective conversation. Help drive
people out of some of these more subjective shadows where they say just trust me, this
is my stuff, I've always done it. Let's put a flashlight on it, take a look at it. That can be
uncomfortable for some people but it ends up in the long run being a more productive
conversation. I think we may have time for one more question. Have to stop at 2:30.
>>: Just to elaborate on that thought. You mentioned two different things. This whole
discussion in the book, it seems that not to fall into the how trap and try to go to the why
and focus on what, but when you relate that to Microsoft, think Microsoft has more of a
how trap, or do you think it's an alignment issue between different businesses?
>> Ric Merrifield: Both I think are true.
>>: Because those are two separate things.
>> Ric Merrifield: I agree. So from a how trap perspective, I have some actual
deliverables that I produced internally where I had the shiny object conversation with
people to say, okay, tell me what you think the five or ten most important things are for
your division. And one particular division I interviewed, I think 44 people in about 50
percent of the time, they name one specific thing as the most valuable thing that could
be done in the organization. Most impactful.
But when I did the heat maps for them and did value and performance maps, that one
thing actually came back as very mediocre, impact both as a child, parent/child
relationship as well as value performance.
So I went back to them and said, you said this was most important. And yet here the
heat map is telling a very different story. And that example they're like, oh, you're right.
These other two areas, they're actually upstream of that one area, it does make sense
for us to focus on those and that will probably fix the one we thought or we said was the
most important thing.
So as an individual discussion, and it turned out in that particular case it was very
common here is people want to throw technology, they see technology as the primary fix
to a specific problem, in a lot of cases it's something upstream of the technology like
better information to go into the technology, better discussion with the customer, which is
maybe something I should have said at the outset.
One of the reasons I've been supported to do this book is it helps Microsoft be better
listeners to customers so United Airlines calls us, we can have that more objective, even
if we're not experts in the airline industry, we can be better listeners to the problem and
say we don't know that much about flight check0in but we can help you step through the
fact to expose it's about reservation, surveys and logistics. Now we can talk about our
best practices there.
So that's the how trap question. The sort of departmental divisional alignment
discussion becomes somewhat of a how trap discussion at a higher level and say, okay,
individual parent/child relationships for specific divisions are one thing. Does the one
thing in this department have the opportunity to have alignment or leverage from this
other area, you can have that conversation more objectively, I would assert, through the
heat maps than you can with process work flows. That would be my assertion. I'll be
happy to talk to you about that off line because I think we're running long. If people need
to be going I'd be happy to answer other questions, though.
>> Kirsten Wiley: Thank you.
>> Ric Merrifield: Thanks very much.