Trend Analysis: methods and problems
Lyn Firminger
March 2003
Trend analysis: Methods and Shortcomings
Abstract
Trend Analysis is used in primarily in marketing, business planning and strategic
planning at both a corporate and national level. It has a number of sub-methodologies;
historical trend analysis, content analysis, cyclical pattern analysis, and the use of
expert opinions such as Delphi. It should not be used as a stand-alone method but can
be useful when balanced with other methodologies. Trend analysis informed by other
Foresight methodologies has the capacity to give analysts a broader and deeper
understanding of possible futures.
Introduction
The futures field and its methodologies are changing. New methodologies are
emerging, and older ones revitalised. This paper explores trend analysis (TA) as a
methodology that has some underlying sub-methodologies, some of which might be
useful in different types of analysis. This is not an exhaustive survey but serves to give
an overview of different methods.
A trend is evident when some phenomenon is seen to have a specified general direction
or tendency. Some trends, for instance those using population data relatively stable,
that is, not likely to be dramatically affected by other events or phenomena, they are
built on other long-term trends such as birth rate, life expectancy and habitation
patterns, which change slowly over time. Trend Analysis (TA) is also used to predict
business, consumer, and political trends at local to global levels. These are less stable
and more likely to be impacted on by other trends or events. For instance, a trend
toward democracy in Indonesia might be indicated by a growing middle class, but this
trend could be slowed or dramatically changed by a rise in religious fundamentalism,
the political „colour‟ of a new president, prolonged drought, regional instability and so
on.
One of the keys to TA is to think broadly and deeply enough about the possible
interruptions to a trend. It is also necessary to try to understand the author‟s cultural
and political perspective and thus their possible blind spots, as TA is not value neutral.
“Pop” Trend Analysis
TA has become increasingly important to both small and large businesses. The
marketing strategy of a business, particularly a small business, is based on finding out
what the trends are in a particular industry, plotting a possible birth to death trend curve,
marking with an „x‟ where the trend is at present (rising, at its peak or falling) and
deciding how much to invest in getting in to that market. This is more true of small
businesses because very rarely are small businesses trendsetters as they do not have
the market power. Much of the development of TA has taken place in the US. At one
end of the spectrum are analysts who operate at a very surface level or who rely on
straight-line trends, without considering interruptions to trends.
One of the gurus of surface level TA in America is Faith Popcorn. Her trend „bytes‟ are
available free of charge on her website. Her products are hour-long seminars,
consultancy and books. Members of the public can join her talent bank for free.
Members supply demographic information, and may be contacted to comment on an
embryonic trend. This is the same logic that is used by marketing focus groups. She
calls her methodology „brailing the culture‟, which is a scanning of popular behaviour.
She monitors the top ten of „anything‟, movies, music, TV, books, plays, products and
fashion. Popcorn‟s particular genius seems to lie in her ability to give a name to
something that is happening among large groups of people so they can begin
discussing and evaluating these trends for themselves, while canny business people
develop the products or services to nurture the trend. For instance, „cashing out‟ is a
trend that involves opting for a simpler lifestyle. I have seen two substantial articles in
popular magazines and heard a number of passing references to this now named
phenomena. Naming it, has given people an opportunity to examine some
phenomenon that is affecting their lives. This type of TA operates at a surface level
and, while it is not likely to give anyone any profound insights into life, it appears to have
its uses.
There are many organisations and „gurus‟ willing to forecast what will happen in the
coming year or decade. These are often business consulting companies that also sell
CDs, books and articles and sometimes free newsletters. One of these is Futurist.com.
The Principal became a professional futurist in 1980 after an academic career.
Unfortunately he does not mention if he did any Futures training. You can look at the
trends for 2004 on their website. The value of this site is that the view is wider than just
economic; it includes quality of life issues too. For instance the “biggest story” for the
year is the declining birth rates, but at least the question is posed - why is this a
problem?
The guru of straight line TA is John Naisbitt who first published Megatrends in 1982.
There have been numerous editions the most recent in 1996. The original TA tool
Naisbitt used was to analyse the content of the news, in newspapers. The premise of
Megatrends (1982) was that „the most reliable way to anticipate the future is by
understanding the present.‟ He did this mainly through monitoring local events and
behaviour „because what is going on locally is what is going on in America.‟ In the
course of this monitoring, he and his colleagues have come to the conclusion that (a)
trends start in local communities not in New York or Washington and (b) trends are
bottom up and fads are top down.
The genesis of this method of content analysis began in the work that he carried out
under the leadership of Paul Lazarsfeld and Harold Lasswell during the Second World
War. They would analyse the content of German regional newspapers to get an
understanding about how Germany was really faring in the war. Information on
supplies, production, transportation and the food situation was secret, but by carefully
reading local newspapers they got a sense of how many factories were being closed or
opened, what their production targets were, and so on. They also found that the local
papers would publish local casualty lists, even though this information was highly secret
at a national level. The military continue to use this type of content analysis. What
Naisbitt did in Megatrends, was to apply it commercially. There is one further crucial
theoretical consideration in his method:
The news hole in the newspaper is a closed system. For economic
reasons, the amount of space devoted to news in a newspaper does not
change significantly over time. So, when something new is introduced,
something else or a combination of things must be omitted... In keeping
track of the ones that are added and the ones that are given up, we are in
a sense measuring the changing share of the market that competing
societal concerns command.
He gives as an example the increasing amount of space being given to environmental
issues and that being matched precisely in column inches by a decrease in news
about civil rights. The trends in this first book were: the information society, personal
value systems growing when potentially dehumanising technological innovations are
introduced, long-term thinking, the decentralisation of power, self-help, participatory
democracy, networked world, the growth in importance of the southern states of
America and multiple options replacing either/or scenarios.
There are a number of problems with this approach. Firstly, the content analysis
devised during the war was not intended as a tool to predict future behaviour; it was
intended to look beneath the news to see what was happening today. Secondly,
newspaper reporting is very selective tending to prefer the sensational to the more
substantial. Thirdly, it assumes that a trend will go through its lifecycle uninterrupted
by sudden shocks. This assumption has Naisbett predict that Japan will replace
America as „the world‟s leading industrial power.‟
Naisbitt‟s first take on TA was very US centred, in subsequent works, looking at Asian
trends he incorporated a view of „out there‟ but filtered through the American press.
Many of his trends have petered out, as one might expect, but he continues to be
successful by replacing them with new ones. Naisbett views progress as linear,
(onward and ever upward), and relatively impervious to external shocks.
A new development - trends and interpersonal development
Just before leaving the „pop‟ arena it is worth noting that TA is now being used to as a
life guide. There are several examples of this. On October 29 2001 an organisation
called LifeCourse Associates asked the question „what will the new normal be?‟ They
argue that America may be entering „the fourth turning‟; that is there is a cyclical
pattern to history evidenced in the past five centuries of America‟s history. Each
„turning‟ takes approximately 100 years. First comes a „High‟, most recently
evidenced in the baby boomer generation, followed by an „Awakening‟ a period of
consciousness, such as happened after the American President Kennedy was
assassinated, with the fourth period being a „Crisis‟ the last of which was the 1930s
depression. This cyclical rather than linear, view of time is more typical of Eastern
culture than Western, and does have relevance for TA, as will be shown below.
Paul Ray uses TA to define a new political grouping, „cultural creatives‟, who are,
according to Ray, „in-front, deep green, against big business and globalisation and
beyond left vs. right.‟ He uses a survey he carried out in 1995 as the basis of his
2002 trend analysis of this new and potentially powerful political grouping. He argues
that the traditional trend analysts, the political pollsters and campaign advisors use,
fails in picking up subtle changes in society.
The methodology is not explained in this paper but I did learn from contacting him that
his sample was 1036 people who after being contacted by telephone (along with an
unspecified number of others) agreed to fill out a long questionnaire regarding their
values. The analysis remains murky but it appears that the only results he saw fit to
publish were those supporting the values he attributes to the cultural creatives.
Despite the use of factor analysis and bell-shaped curves this trend sub-methodology
ultimately fails to convince because the sampling and analysis are a lot less than
transparent and American politics is far more complex than is made out here. This
type of TA purports to look beneath the surface but relies on selectively picking
through history, leaving behind what is not appealing and using it to support a thesis
(the existence of „cultural creatives‟) that in itself is analytically on shaky ground.
Trend analysis as a business tool
There are many more seriously presented TA consultancies, some are industry specific,
others such as the Australian company IBISWorld cover all industry sectors in some
depth. In Australia the need for business intelligence utilising trend data increased in
the 1980s.
During the 1980s there was a large increase in micro businesses, particularly in the
service sector. The influx of businesses meant that a business had to link into trends
to survive. During the following 10 years the business environment became even
more chaotic, harder to predict. One consequence was that a raft of consultancies
emerged to assist businesses identify trends. One, important and highly respected
such organisation in Australia is Phil Ruthven‟s IBIS World. It collects information on
industries using hard data sources such as industry turnover, employment, number of
establishments, etc as a grid to plot rises and falls in the fortunes of that industry. This
is supplemented with qualitative data; newspaper stories on companies in the industry,
annual reports of public companies, etc. This information is then sold.
Non-subscribing visitors to the website can get a one-page overview for free and regular
readers of Business Review Weekly will get the encapsulated form of the trends, as
there is a tie-up between the two organisations. The information can also be sourc ed for
free at Victorian Government Business Centres.
McKinsey and Company are a large US based, international Consultancy that work with
larger businesses to develop business strategies. They also publish an online
newsletter with some free information, and a quarterly journal. They will follow a
number of „hot‟ trends over a period of time across a wide range of industries. There
are many hundreds of consultancies offering TA as one of a suite of tools in business
planning. Some purport to take the uncertainty out of business planning, but it is difficult
to see how this might be possible using the sort of linear trend projections they appear
to advocate. Several serious trend analysts have published „how-to‟ books on TA and
these can assist unravel some of the mysteries.
Trend Analysis as a strategic planning tool
Gerald Celente has written his book on tracking trend with an aim to profit from them.
It begins with endorsements, mainly from newspapers and gains credibility by being
distributed by the World Future Society. It claims to „go beyond the rosy scenarios in
Megatrends 2000. The Institute of which he is a part started tracking trends back in
the 1980s. This book has been written in the belief that most people react to trends -
and therefore fail to benefit from them - whereas smart people „proact‟ in the face of
trends. That is, they „anticipate the future and act accordingly hopefully ahead of the
one to three year timeline business and government use.‟
The sub-methodology used by Celente is newspaper analysis, but at a more
sophisticated level than Naisbitt. He goes into some detail about how to read
newspapers and mark them for future reference when you are reading huge amounts
of information. Trends come from different groups in societies as well as from social
leaders. It is important to have a „search image‟ when you are searching, that is one
or more trends you are looking for, but these must be broad enough to encompass
trends you might not have thought about to begin with. He illustrates this with a story:
if you go into the jungle just looking for monkeys, you will not find snakes. Implicit in
this is that if your expertise is too narrow, you may well miss something important. He
argues that to be a trend, an event must „follow logically ... from causes that span a
number of fields or disciplines (and have) ... at least two events to form a sequence.‟
The fields the trend must appear in are social, political and economic. Trends will
have a long life and go through the phases of birth, maturity and death; fads on the
other hand are short-lived and without social, political and economic significance.
Celente is aware that if you are going to use newspapers you need to be
discriminating in your use of them, he does not, for instance, use content analysis
because so much in the newspapers is unimportant. This is considerably different to
Naisbitt with his unquestioning acceptance of all the news that is printed within the
news hole in the press. Celente gives a detailed description of his method:
You have to screen out the headline news and the junk news and
focus on the real news, the current events that are forming future
trends. Real news rarely appears on television and it rarely makes
headlines. It is usually in the middle of the better newspapers or in
specialized publications. There you can find warnings of a crisis
months or years before it makes headlines.
The second part of the book is taken up with looking at some major trends; these are
media, politics, the family, education, health, the environment, the military and the
world. These areas are treated, naturally enough from an American perspective. It is
interesting to look back at this 1980s view of what the trend drivers were: he sees a
trend to globalisation as a move away from ideology. The new century would leave
behind the ideologies of communism, fascism, Nazism, and fundamentalism.
Standing in 2004, we can say this hasn‟t happened yet.
Michael Mazarr weighed into the trends debate with a book looking at trends to 2005
but with a proviso that we have around ten years to influence these trends so that the
more positive rather than negative aspects prevail. He also takes a broad world view.
He is involved with an organisation CSIS, the Centre for Strategic International
Studies. Financial support for the book came from The McCormick Tribune
Foundation, The Korea Foundation and the BP-AMOCO foundation. Knowing this
assists in understanding his biases.
He takes his starting point from Peter Drucker:
Every few hundred years in Western history there occurs a sharp
transformation. Within a few short decades, society rearranges
itself - its worldview; its basic values; its social and political structure;
its arts; its key institutions. Fifty ears later, there is a new world.
The current transformation will affect the West and most of the rest of the world. The
transformation is from an industrial to a knowledge era and will be driven by science
and technology but social, political and psychological changes will follow and
complement these driving changes.
He draws on a wide variety of sources from Daniel Bell‟s 1970‟s work on
post-industrial society, Charles Handy‟s notion of paradox (which permeates the book)
and the evolution from modern to post-modern society, John Peterson‟s
wild cards, and Paul Ray‟s cultural creatives, among many others. Mazzar discusses
the idea of moving from a Newtonian image of the predictable universe toward the sort
of holism evident in quantum mechanics and complexity theory. His inclusion of
quotes from Tao Te Ching suggests a willingness to look at cultures other than his
own. The book does not contain an explanation of the methodology or prescriptions
for trend analysis, but it is evident that it derives from very wide reading. In the
substantive areas of the book it is clear he draws on United Nations data, media
stories, and many varied writers.
The book is problematic in that it has an overly complex layout and the Chapter
headings often do not give a sense of where one is heading. Nine themes underlie
the book. Issues dealt with in the book include world population, population and
immigration, the authorship of the knowledge era, knowledge and the reorganisation
of work, and globalism, tribalism and pluralism.
In the conclusion he remains optimistic we will overcome problems, and that there will
be a synthesis that grows out of whatever crises accompany some of these changes.
One of the areas of synthesis is between Western and Eastern worldviews. Neither
will predominate but there will be „mutual absorbsion‟ between the two. Mazarr is
much more reflective than other analysts and his book could provide a useful tool to
help a researcher think through the various issues.
Trend Analysis and multiple futures
The different TA methodologies discussed so far have been predictive to some extent
trying to pick the trend or the future of something. This need not always be the case,
TA can be used to set out different possible scenarios.
Barney Foran and Franzi Poldy, two CSIRO researchers undertook a study into the
impact of the size of Australia‟s future population on the environment, the physical
economy, the national infrastructure and our quality of life. They used trend data
going back some 50 years and then extrapolated this data to 2050 using three
different population scenarios; high, medium and low population growth. From this
they distil a number of possible dilemmas Australia may face in future. An appendix
gives comments from the reference group. This is an important work and if the
Government make a choice as to the population scenario to opt for, will be a useful
tool for those involved in demographic research.
The US Government has considerable resources to fund futures research. One such
piece of research was published as Global Trends 2015: A dialogue about the future
with non-government experts was the second such report developed by the National
Intelligence Council. It was approved for publication by the National Foreign
Intelligence Board under the authority of the Director of Central Intelligence. It looks
at the world „from the perspective of the national security policymaker.‟ It engages with
outside experts from academia and private organisations rather than relying on
traditional classified sources of information. The project began in autumn 1999 and
was published in December 2000 and builds on an earlier work published in 1997.
The work identifies global drivers and estimates their impact on the world over the
next 15 years - demography and natural resources, technology, globalisation and
governance, likely conflicts and prospects for international cooperation and the role of
the United States.‟ This is a large broad-ranging document with a focus on American
security, as one would expect from its authorship.
The process allowed for a rich informational input. There were workshops attended
by Government and non-government specialists from a wide range of fields. Each
workshop had a different focus, one „demography, natural resources, science and
technology, the global economy, governance, social/cultural identities and conflict and
identified main trends and regional variations.‟ The second workshop was a scenario
planning exercise built on the information from the first workshop. It highlighted „key
uncertainties, discontinuities and unlikely or “wild card” events and (identified)
important policy and intelligence challenges.‟ Thirteen conferences were held; each
co-sponsored by the NIC and other government and private centres. Each had a
different theme, for instance evolution of the nation state, trends in democratisation,
American economic power, and alternative futures in war and conflict, to name a few.
A draft report was then put together and circulated to outside experts and the final
report developed. The paper begins with drivers and trends, looks at key
uncertainties and then key challenges to governance. The next „discussion‟ section
covers the substantive areas of population, environment, etc. This is followed by a
thumbnail sketch of the major regions. The four alternative global futures are
contained in the Appendix along with a matrix of the drivers (population, resources,
etc) and how they might develop under each different scenario. While this work has a
focus on the US, its underpinning information can be used to analyse trends in other
countries too.
The changing methodological paradigm
There is a growing need for TA and other Foresight methodologies that are capable of
allowing for multiple possible futures. In some cases these can then act as
organisational discussion starters about preferred futures or as input into Scenario
Planning exercises. Drivers of this change include growing global uncertainty the
shortening of the business cycle and a shift from an industrially based society to one
based on knowledge and information. Underlying the shift in Futures thinking is the
decline of Positivism as a philosophy underpinning social discourse and the rise of
modernism and post modernism.
A recent survey of the life work of Wendell Bell shed some light on this changing
futures paradigm. For Bell an assertion of a possible future - or anything else - must be
accompanied by evidence, the assertion must be able to stand up to the rigour of
attempted falsification of the hypothesis. Bell‟s intellectual roots are in the American
„scientific‟ sociological movement. Bell has acknowledged that critiques of positivism
have helpfully corrected an uncritical acceptance of positivist science‟ but if taken to a
logical conclusion „“not only would causality, determinism, necessity, objectivity and
rationality be abolished, but also humanism, liberal democracy, responsibility, truth itself
and, we can add, futures thinking.”‟ TA has embedded into it the sort of conjectural
knowledge that „incorporates justified beliefs about the future‟ typical of this tradition
according to Stevenson.
Is TA doomed to remain within the positivist or critical realist paradigm as Bell calls his
move beyond positivism? Does it have any relevance in Future Studies today? TA
certainly suffers from all the shortcomings Richard Slaughter found in his critique of the
Futures domain as it stood in the early 1980s. It tends to be superficial, „it fails to
recognise the roles of language, power and embedded social interests, it lacks
understanding of its own sources and grounding‟, it is appropriated by the powerful, it
suffers „overconfidence in easy prescriptions‟ and it is not open „to other ways of
knowing.‟ Slaughter is concerned with the emancipatory role of futures, its role in
reconnecting the inner world (not taken into account in the positivist tradition) with the
outer world.
Rescuing Trend Analysis for the Future
Trend Impact Analysis (TIA) is a methodology that explicitly deals with unexpected
events as part of a quantitative based TA. The method allows for the systematic
treatment of possible future events, be they technological, political, social, economic
or value-oriented.
Expert judgments are sought about the probability or an event as a function of time
and its expected impact on the trend under consideration. To begin with, historical
data is selected and a surprise-free extrapolation is developed using a curve-fitting
algorithm, as used by Foran & Poldy. At this point the judgment and imagination of
the researcher is critical. A list of possible potential events is prepared. „These
events should be plausible, potentially powerful in impact and verifiable in retrospect.‟
The source of such a list might be a Delphi study, some form of other informal
consensus among experts or a literature search. The probability of the event is
estimated then an estimate is made of when (or the duration of) the start of the effect,
followed by the impact of the trend at its largest or when the impact might reach a final
or steady-state level. Alternatively one might try to estimate the magnitude of the
event the largest impact or the steady-state impact.
The paper discusses TIA software that gives a quantitative analysis of probability of
occurrence and the impact on the trend. This may be useful to Managers wishing to
estimate the probability that strategic targets might be met over a number of years. A
simple to follow example of the continued use of chlorine in bleaching paper is used
as the example. Gordon assesses the strengths of TIA as that not only does it allow
you to say “Here‟s where I think the variable will go in the future,” it also allows you to
add, “and here are the events I‟ve taken into account.” He likens this to constructing
a scenario, just as the NIC did in their study. A further advantage of TIA to the
strategic planner is that once the possible events are identified they can be built into a
scanning framework for the trend.
A number of other Futures techniques might also be used to enrich TA. Wild cards is
one, there is an increasing recognition that the wild cards developed by John Peterson
need further development to be of continuing use to those outside America. Mark
Justman took Jim Dator's four images of the future continuation, crash, slowdown, and
transformation; and mapped them against Peterson‟s wild cards to enrich their
possible usage. As we have seen above Delphi and Scenario Planning have also
been used to enrich TA.
Conclusion
Trend Analysis is particularly useful in marketing, business planning and strategic
planning. To carry out TA a researcher will probably have carried out an
environmental scan and detected that there appear to be trends in some item of
interest. Something is continuing to hold an ascending or descending pattern or when
A happens B almost always follows. The sources of information used to follow up on
the trend will vary with the trend under consideration. Sometimes news appearing in
the tabloids may be a good way to follow a trend, usually other methods will be
needed. Some of the analysts surveyed here draw on surveys, historical analysis,
conversations with experts, and so on. The difference between John Naisbitt and
Michael Mazarr for instance lies in the diversity of inputs into the analysis. The
critical question to ask at every juncture is „could anything happen now or at some
point in the future to disrupt this trend?‟ Follow this with questions about the likelihood
of an event occurring and its possible impact. Ask these questions through various
experts; this could take the form of a literature search, an expert panel or a Delphi
panel. Explicitly factor the results into the TA; the outcome will be that the researcher
will have a broader understanding of what is happening now and the possible impact
on future events.