Trends in Food Science & Technology 50 (2016) 243e248

Contents lists available at ScienceDirect

Trends in Food Science & Technology

journal homepage: http://www.journals.elsevier.com/trends-in-food-science-
and-technology

Viewpoint

New product failure: Five potential sources discussed

Garmt Dijksterhuis a, b, *
a
Sensory and Consumer Science Section, Department of Food Science, University of Copenhagen, Rolighedsvej 30, 1958 Frederiksberg C, Denmark
b
Utrecht University, University College Roosevelt, P.O. Box 94, 4330 AB Middelburg, The Netherlands

a r t i c l e i n f o a b s t r a c t

Article history: When new consumer products are developed and later launched, 50 to 75 percent of them are removed
Received 18 February 2015 from the market far short of meeting their projected financial targets. In short: they fail. We conclude
Received in revised form that this failure is due to institutionalized insufficiencies in the use of the sciences that are best geared to
18 January 2016
understand and predict consumer behaviour, viz. the behavioural sciences. These are not necessarily the
Accepted 26 January 2016
Available online 1 February 2016
same as the marketing science that is performed by marketing departments. A scientific approach to
understanding consumer behaviour appears to be lacking in many corporate research surroundings. This
often is in great contrast with their high levels of technological science, paralleled by their respective
Keywords:
New product failure
research budgets. In this paper we present five problem areas that may contribute to this mismatch,
NPD contributing to needlessly high numbers of product failures. We have termed these factors: (1) ‘pillars’
FMCG (too many different functions addressing different aspects of the consumers and of product develop-
Industrial research ment), (2) ‘higher management focus’ (not geared towards understanding consumer behaviour), (3)
Behavioural sciences ‘popular science books’ (out-dated research directives resulting from a hierarchical management model),
(4) ‘quality and Quality’ (a definition of ‘quality’ that leads to invalid quality parameters), and (5) ‘psy-
chophobia’ (the latent fear of trusting behavioural science results), respectively.
© 2016 Elsevier Ltd. All rights reserved.

1. Introduction Even though a ‘consumer focus’ was introduced as a guiding

Most people in the FMCG business, and in particular in the food
area, agree that the fail rate of new product launches is too high.
The Nielsen breakthrough innovation report (Nielsen, 2014) has
analysed 12,000 new FMCG products that were introduced be-
tween 2011 and 2013 onto the Western EU markets. They conclude
that 76% did not last one year of sales. What's even more shocking is
that according to their analysis 45% did not last 26 weeks. In an
older study, in the USA only 53.3% of new products is reported to
achieve their financial objectives (APQC, 2003). The exact numbers
are a matter of debate and will depend on the definition of failure
and on the specific area of consumer goods. Nevertheless, in FMCG
most would accept a rather high number, somewhere between 50%
and 75%. Some people appear to regard these levels of failure as a
law of nature, which of course they are not. They rather instead are
a painful reminder of the fact that most new product developers are
devoting considerable time and money to products that should
have revealed their flaws at the moment of inception or early stages
of development.
* P.O. Box 94, 4330 AB Middelburg, The Netherlands.
E-mail addresses: g.dijkster@ucr.nl, garmtdijk@gmail.com.
principle in new product development some twenty to thirty years
ago, and marketing departments aid decisions on the direction that
R&D should take, still we are confronted with that too-high fail rate.
Over the years we have identified a number of underlying issues
of corporate R&D culture that may disproportionately contribute to
the development of unsuccessful products, which we will present
and illustrate in this paper. We name them:
1. pillars,
2. higher management focus,
3. popular science books,
4. quality and Quality,
5. psychophobia.
We would like to point out that our experience does not stem
from any particular product area, industrial branch or geography.
We collected our experiences over many years, in a wide range of
areas, albeit mostly in the FMCG and food areas. Our interactions
with fellow workers in the field ranges wider than FMCG, and in-
cludes many different types of consumer products and professional
corporate and applied research environments.
We present our arguments against the background of new

http://dx.doi.org/10.1016/j.tifs.2016.01.016
0924-2244/© 2016 Elsevier Ltd. All rights reserved.
244 G. Dijksterhuis / Trends in Food Science & Technology 50 (2016) 243e248
product development, but they are probably equally true in other
areas where consumer products are adapted, conceived of, mar-
keted, etc.
2. Pillars
Most, if not all, of the industrial scientists in research, marketing
or other related functions, that we encountered over the years
agree with our statement that the way their research environment
is organised is very conducive to working in pillars. Individual
workers in R&D can often see the wider context of their work, but
just as often they see it destroyed by inadequate communication
between pillars. There are simply too many different functions. In
some organisations we find separate functions for research, pro-
duction, marketing, sales, customer, retail, brand, delivery, to name
a few. For each function there is a separate department, with their
own directors, targets, budgets, language, sometimes in their own
geography. This very often hinders efficient and honest commu-
nication between such departments, sometimes even to the point
of hostility.
At the same time we sometimes witness very different science
areas lumped together in the same pillar. In some cases behavioural
scientists sit in teams with microbiologists or physicists, to work on
the same development. Although a story that very broad teams
work on the same problem may sound good to higher manage-
ment, in practice there is often too little synergy at the level of the
actual project content. Workers in such (overly) multidisciplinary
teams are either too nice or too hostile to other disciplines, and may
lack the culture that enables them to dispassionately give and
receive, and act on, critical review and constructive challenges. The
result is that challenges are seen as a threat rather than a crucial
step in reducing fail rate. A more optimal grouping of research
problems and distribution of research scientists over problem areas
is certainly possible in many such cases, as is a culture and mind-set
that fuels synergy rather than discord.
Some breaking down of pillars can be seen in P&G's ‘growth
factory’ model (Brown & Anthony, 2011). In this model a new
organisation is built that aims to ensure increased communication
between some of the pillars that could have slowed down inno-
vation in more traditional models. Some of the success of recent
innovations of P&G are claimed to be the result of this new model.
3. Focus of higher management
A consumer-led research agenda is not the same as a marketing-
led one. Very often the development of new products is guided by a
marketing function. However, the horizon of most marketing de-
partments is not very deep. They are geared to launching a product
next week or next month, while R&D needs a horizon of years to
develop the radical new insights that are expected of them. Steer-
ing them from a marketing point of view can only lead to mediocre
innovations. Furthermore many marketing managers are on a
rotation scheme, rendering it less likely that they are confronted
with the failures of the products they launch, because by that time
they are in another job. There are also many career-marketers,
favouring their own career over the ‘career’ of the products they
develop. Product marketers are not always research minded, and
their surrounding department may not welcome a scientific atti-
tude. When behavioural or consumer science is applied, these
managers often fall back on the standard methodology that they
learned as students. However, those methods are likely to be un-
sophisticated and outdated, and are rarely critically reviewed in the
light of the latest scientific insights in psychological science.
3.1. External expertise
Related to the previous section is a reliance on external exper-
tise. Sometimes this is expertise that the company bought-in years
ago from a convincing consultant. Agencies offer their expertise,
but often they sell their own methodology based on their own,
possibly idiosyncratic, ideas about consumer research. Many of the
standard methodologies offered may be suboptimal with respect to
the current research problems. Only a few agencies will develop
bespoke research methods and display a critical attitude towards
their own findings.
In this context the quote of Henry Ford, the automobile entre-
preneur, about consumer research is telling: “If I had asked my
clients what they wanted, I would have been breeding faster
horses.”. Many new product developers appear to be in the busi-
ness of breeding faster horses, resulting in a research program that
is running within a narrow window of minimal improvements to
the existing ranges, rather than focussing on step change innova-
tion. They are caught up in tweaking the details of products that are
already near perfect for what they are. Innovation has become
entirely focused on how best to deliver the current product, rather
than how best to deliver the benefit consumers seek from it. The
risk is that one may become blind for the type of radical innovation
that the competition is working on, and is taken by surprise when
the competition launches a game-changing alternative product.
The reaction “Why haven't you thought of this!” can then be heard
in R&D departments, where higher management subsequently
blames the research scientists whose hands they themselves tied
by the limitative directions to ‘breed faster horses’.
3.2. Budgetary imbalance
Not unexpectedly, many or most R&D leaders have a techno-
logical background. As a result most of the work will be devoted to
product technology, and research on the consumer as a psycho-
logical entity may be snowed under in the many possibilities that
new technological developments offer. The focus of research de-
partments thus led will be on finding new molecules, formulations,
ingredients, processing, etc. There will be a movement towards the
use of the latest equipment and the scientists will be caught up in a
race for patents and in beating the technology developed by the
competition.
As a result of this a severe imbalance arises between the re-
sources for behavioural and for technological research. A twenty-
fold bigger annual research budget for technology compared to
behavioural science would not be unusual. No wonder that the
really big innovations occur in the product field and never occur in
the consumer field. Yet these product-technological innovations
often lead to unsuccessful products because consumer-relevant
aspects are overlooked or simply never sufficiently addressed.
When comparing the cost-effectiveness of technological to
behavioural research, the latter is likely the more cost-effective of
the two by far. The costs of behavioural studies are relatively
modest, and they give valuable information. The costs of per-
forming standard consumer tests of a new product, even for a large
sample of one thousand consumers, are small compared with the
costs of a product failure. For step change innovations in the un-
derstanding of consumer choice, much larger sums are needed, but
they will likely not exceed the requirements of most technological
studies where very expensive equipment and materials are needed.
3.3. Open Innovation
Another matter of concern is a new model for industrial coop-
eration: Open Innovation (cf. Chesbrough, 2005). In principle this is
 G. Dijksterhuis / Trends in Food Science & Technology 50 (2016) 243e248
a good idea, but the increased reliance on external experts also
holds a risk. If external expertise is sought, at the expense of in-
ternal expertise, there may be not enough internal expertise left to
critically evaluate the available external expertise. Up-to-date in-
ternal scientific expertise is needed to guard against the over-
promising by external scientists. Universities are increasingly
pressed to valorise their research so they are not inclined to say ‘no’,
to industrial requests. In addition universities have a different
definition of ‘project success’, than industries have, easily leading to
mutual disappointment and a waste of money. In particular, for the
social/behavioural science departments of universities cooperation
with industrial R&D may be a totally new venture.
Usually external services are pitched to and evaluated by senior
R&D managers who have not actually practiced science for many
years, often to the specific exclusion of their own (junior) experts.
Commercial agencies and consultants are prone to oversell their
products, and may be using older technology that is not recognised
as such by a lack of internal expertise. An example in the behav-
ioural sciences is the broad application that qualitative consumer
interviews still have, or claims for finding the ‘buying button’ in
consumers' brains, in some neuro-marketing applications.
Most OI models are developed for technological R&D and not
directly suited for use in behavioural and social science research.
The results of technological R&D can be expressed in the delivery of
a technological item (machine, molecule, process), that the in-
dustry may subsequently license or own as part of the agreement.
The findings of behavioural research are not of this ilk, and may not
fit the thinking of most OI models.
4. Popular science books
Some form of hierarchical management model is still prevalent
in most industrial research surroundings. The result of this is that
whenever a director, CEO or VP launches a new insight or idea, it is
not openly questioned by the lower echelons where the research
scientists live. The higher management are not behavioural experts,
nor do they operate at the forefront of the developments in con-
sumer science. They reflect on what they pick up from their peers,
and what they may read in popular books and media on behav-
ioural science. Now, reading such material is commendable, but it
does not serve you if your responsibility is to create a competitive
edge. Take as an example the popular psychology book “Thinking
fast and slow” by Daniel Kahneman, published in (2011). Nobody
would question the great scientific value of the psychological sci-
ence by this Nobel laureate. But when our industry leaders read his
book and conclude we have to work on heuristics, just to name an
example, we are in fact 35 years too late. The work on prospect
theory that lies at the basis of much of Kahneman's work was
published in 1979 (Kahneman & Tversky, 1979). If we wanted to
create a competitive edge in understanding consumer choice based
on that work, we should have done it back then.
This means that if we want to be at the forefront of under-
standing consumer choice, we should listen to, or create, the
behavioural science of tomorrow today. It does cost, but it will
repay itself, and the investment costs are usually much smaller
than those in the technological sciences. Industrial behavioural
scientists can create new science, tailored to the specific industry's
needs, if only they are seen on an equal footing with their tech-
nological peers in the lab. We know that their standing as expert
behavioural scientists is not at stake, but it hardly ever translates
into budget or into a voice creating a longer term vision on the
direction that industrial consumer science research has to take.
In the field of behaviour change some developments are taking
place in the direction of a health oriented behaviour, where an
FMCG industry chose to, or is legally forced to, lower the levels of
245
salt, sugar and (saturated) fat in their products. Maintaining con-
sumer choice for these products creates the need to understand the
determinants of consumer food choice. Psychological principles as
implementation intentions or nudging are being investigated to
this end (see De Ridder, 2015). Nudging (Thaler & Sunstein, 2009) is
another ‘popular’ method of changing consumer behaviour into a
certain direction. The theory behind nudging is much older and
goes back to priming (cf. Schacter 1987, Bargh, 2006) and framing
(cf. Kahneman & Tversky, 2000).
5. Quality
A phrase often heard in R&D surroundings is that ‘quality
products’ are what we need to develop, because consumers will
buy them. Here reference is being made to ‘intrinsic product
properties’, or the ‘product itself’, as if quality is something that
resides in the product. It does not, and the definition of quality that
many product developers seem to use is often phrased in a topsy-
turvy way. Many would agree with the following statement: High
quality products lead to good sales figures. This seems to be the
dictum much product development operates under. We do not
agree to this statement, and instead pose: Good sales figures lead to
high quality products.
Think about what a definition of ‘good quality’ for any consumer
product should look like. In an industrial, thus commercial, envi-
ronment there is only one thing to base a definition of quality on:
sales! If initial consumer take-up is satisfactory but a product does
not maintain repeat sales, then its quality was insufficient. If a
product achieves high sales figures, it thereby has great quality. You
can argue if it is sales figures, ROI, profit, or any other economic
measure, but this is what should define quality. Quality is often not
what developers think a product should be. If a marketer is heard to
utter: “I don't know why this product does not sell, all its quality
parameters tick the box.”, this is a sign of failure of the consumer
research and marketing functions. The answer is obvious: the
quality parameters are incorrect. By definition. The lab-tests for
quality need updating. Quality standards need to be updated based
on products' market success.
As an illustration, Fig. 1 presents a view on quality (lab tests for
quality) and Quality (based on product success in the market)
where the most important part of the process is the gauging of the
(lab) product quality test using information from the market (be
they sales figures, results from test markets or consumer tests).
In other words, based on variation in market success it should
not only be possible but essential that consumer research tools and
quality parameters are continually challenged and revised to reflect
reality rather than past practice.
6. Psychophobia
This may be the more contentious section of this paper. Clearly
the scientific developments in understanding eand predicting-
consumer choice of new products have to be expected from the
sciences that deal with human beings: the behavioural sciences.
These typically are psychology, sociology, economics, anthropology,
and ethology. We refer to them here as the ‘behavioural sciences’,
but one could also refer to them as ‘experiential sciences’ as they
aim to understand eand predict-the subjective experience of
consumers (cf. Dijksterhuis, 2014a).
When explicitly asked, higher R&D management will always say
they value the work of the behavioural scientists in the lab. How-
ever, when judged from management behaviour this is not always
so obvious. Apart from the aforementioned issue of resources, it is
no exception that departments with behavioural scientists are
usually also led by a person with a technological rather than
246 G. Dijksterhuis / Trends in Food Science & Technology 50 (2016) 243e248
Fig. 1. Quality (capital ‘Q’) model where consumer test, test market results, and other field tests are used to gauge the lab tests that define quality (small initial ‘q’). This gauging
ideally takes place on a continual basis. We here abstain from explaining the details in the figure, it here serves as an example only.
behavioural background. This person will need several years to
grasp the type of research that behavioural scientists do, the tools
and budgets they need, and the type of conclusions that can and
cannot be drawn from their research. This situation is not uncom-
mon, and often leads to friction, suboptimal results of research,
wrong directives for developments and even shrinkage of a
department. If the latter was not the intention, this is a waste of
energy and money (even despite the small budgets involved).
Difficult working situations are prone to arise, and the department
leader is unable to adequately represent his/her colleagues in the
higher management. Yet at the same time, simply by virtue of
position, he/she will often be seen by the company as ‘the’ expert in
the behavioural sciences. We have heard reports about behavioural
scientists of such a department being accused of producing mere
‘psychobabble’. The source of the misunderstandings are not
sought in the fact that the types of research done in behaviour and
in technology are completely different, but rather the behavioural
scientists themselves are made feeling responsible for their ‘failure
to communicate’. The analogous situation, a technological depart-
ment being led by a behavioural scientist, is never seen and never
have we witnessed someone being accused of ‘technobabble’. In a
healthy working environment both terms should not be used. The
asymmetry presented here clearly points toward a latent psycho-
phobia by higher R&D management.
7. Examples
In this section four examples are presented that we believe may
have resulted from the fact that much industrial behavioural sci-
ence is an underdeveloped and underutilized area. Although these
examples may have originated in an FMCG context they contain
lessons that can be easily generalized to other areas.
7.1. Blind testing
In section 5 the problem of ‘quality’ (and ‘Quality’) was intro-
duced. A wrong (not valid) definition of quality will lead to lab tests
being performed that are not indicative of Quality, i.e. from a con-
sumer point of view. The idea that ‘Good quality products lead to
good sales figures’, will result in lab tests aiming to find that quality
in the product. One of the results of this definition of quality is the
type of testing that is called ‘blind testing’. The idea is that the
‘product itself’’ must be ‘good’ in order for it to be sold, and that
when a product can be shown by lab tests to possess ‘high quality’,
it can be confidently marketed. However, a consumer will never
encounter a product ‘blindly’ Never will a consumer experience a
product without even the slightest idea about what it is he or she is
consuming. There will always be at least a little (and usually quite a
lot of) top down information that goes with the product. Such in-
formation has been shown to affect the perception of the product
(cf. Woods et al. 2011), and it will be seen by the consumer as an
integral part of the product (cf. Dijksterhuis, 2012, 2016). The
product will also be remembered together with this information,
and together with the situation in which it was encountered and
consumed. A blind test does not add this information, so the
product in a blind test is only a small part of what the product is for
a consumer. Hence there will always be an incommensurability of
blind lab tests with consumer tests and market success of a product.
Blind tests may also result in directions for new product develop-
ment that are not actually related to current consumer wishes or
can contribute only trivially to overall product acceptance and
sales. The results can thus be poor targets for investment in tech-
nical R&D innovation or, worse, a new product that will not be
accepted by consumers, and will help raising the fail rate of new
products.
7.2. E-Phobia and clean label
These days much research activity in food product development
is paradoxically aimed at the removal of specific ingredients from
products, even where these ingredients are known to contribute to
product attributes highly desired by consumers. The reason is to
circumvent having to name them on the ingredient declaration. In
particular substances that have to be declared on the label by a
 G. Dijksterhuis / Trends in Food Science & Technology 50 (2016) 243e248
‘chemical’-sounding name or an ‘e-number’ (a code used in place of
a name for a range of food ingredients in the EU) are targets for
these activities. The reason for this is that there is a belief that a
very significant number of consumers are put off by the presence of
these in products.
The evidence for the existence of this belief mainly comes from
flawed research where an e-number or a specific ingredient is the
only thing the research focuses on. Consumers in such research
could be asked: “Can you tell me if you like the e-numbers on this
label?” or “Do you prefer products without e-numbers?”. Apart
from the fact that asking such direct questions is bad practice for a
number of reasons, the narrowing down of the food-world to only
food with or without e-numbers puts a stress on the importance of
the e-numbers for this question. There is no room for other parts of
the food situation to enter the judgment of the consumers in such a
study. As a result the responses of consumers in such studies will
evolve around e-numbers and ingredients, despite the fact that in
real life, in real food choice situations, this may only account for a
few percent points of explanation of their food choice. Hard(er)
scientific evidence for the importance of such a belief of consumers
within the context of other product attributes is not abundant.
What is more important, hard scientific evidence for the effect that
such a belief has on the buying of food products by consumers
appears to be lacking at all. With ‘scientific evidence’ we mean
experimental studies or studies on test-markets. Consumer reports
in interview situations do not count as ‘scientific evidence’, as it is
known to be not very predictive of consumer behaviour in the
market place (cf. Ko € ster, 2003). A recent study into the effect of food
ingredient labels has indicated that this effect may indeed be small
or even non-existent (Cheung et al. 2015). If the worry is about
consumers (not) buying food with e-numbers in them, this very
behaviour is what should be studied in experiments. This means
setting up special choice situations where consumers make choices
in as much a natural setting as possible. It does not mean inter-
viewing consumers about e-numbers. The latter type of study is of
course much cheaper than the former, but the money lost in
launching products that are not bought (or in this case, money not
gained in not launching products that would have been successful),
will likely outweigh the research costs by many multiples.
Large sums of research money are spent to remove or replace
specific ingredients from products, based on rather flimsy market
research. This is an example of a directive for research that can be
seen to result from not performing valid behavioural research. We
have spoken to many food (behavioural but also technological)
scientists in many industrial labs, and many of them agree that the
aim for empty ingredient labels may be a case of misguided
research. However, they will continue working on this subject as
their criticism is not heard by higher management, not taken
seriously, not appreciated or it is even never uttered out of fear it
may compromise the scientists' position in the research
organisation.
7.3. Implicit messages from ingredient replacement
A clear case of lack of consumer psychological insight is visible
in the way some industrial research is investigating the use of
specific ingredients or new production processes. This case applies
to artificial sweeteners, GMO's, irradiation, nano-technology and
also to the above mentioned e-numbers. For all these examples
manufacturers and regulators (and increasingly even consumer
advocacy organisations) accept there is no issue of safety for
approved ingredients and processes. Nevertheless some industries
devote large sums of research money to removing or replacing the
ingredients or processes by alternatives. This is entirely based on
the wrong belief that some consumers have. The message the food
247
industry gives by this replacement is counterproductive and will
add to their bad image. This implicitly communicates the message
that indeed there is something wrong with artificial sweeteners/
GMO's/irradiation/nano-technology/e-numbers, and that they
should not be used in food. Consumers will reason that if it were
perfectly safe there would have been no reason for the industry to
remove it. The removal thus feeds the emotional appeal from
consumer groups that believe artificial sweeteners/GMO's/irradia-
tion/nano-technology/e-numbers are a bad thing, and that the in-
dustry is equally bad by having allowed these in their products in
the first place. This negative image will generalise to other products
and possibly other industries. More knowledge of consumer psy-
chology and research into the actual effects of the removal of in-
gredients (in terms of actual consumer choice and longer-term
product perceptions) could have prevented this situation.
This case illustrates the fear that the food industry has for the
consumer, based on a limited understanding of consumer behav-
iour. Marketing studies do not provide the basic insights in con-
sumer choice, for reasons outlined above. Basic behavioural science
and its application is lacking or lagging behind technical scientific
developments, possibly for all the reasons we have mentioned.
A related mistake originates in taking such consumer criticism
too seriously. The number of consumers actively voicing their
concerns about artificial sweeteners/GMO's/irradiation/nano-
technology/e-numbers is relatively small. Furthermore they are
likely to be an ‘elite’-consumer group, with good communicative
skills and the internet and media networks to propagate their
opinion. It may well be the top few percent of a Gaussian curve with
concerned consumers. Having your research programmes deter-
mined by those displays a grave (and expensive) misunderstanding
of the sociology and psychology of consumers and their behaviour.
7.4. Authenticity, naturalness etc
Another movement is the drive for ‘authentic’, ‘natural’, ‘arti-
sanal’, etc., products. Much research into this direction is motivated
by the fact that many consumers use these words when they are
asked about what they find important in products. Use of these
words however does not automatically mean that there is a relation
between these words and what consumers buy. Many current
research projects are run to find ways of making products appear
more ‘natural’, ‘authentic’ or ‘artisanal’. Often consumers are asked
what products they find have these properties. Relying too much on
consumers' explicit utterances is a sub-optimal strategy, resulting
in research money being spent in projects that may not result in
products that are bought by consumers at all. Few behavioural
studies are executed to test the effects of the use of such terms on
consumers' actual choice. Studies on what consumers report in
response to these terms are legion, but they are probably lacking in
validity with respect to consumer behaviour in the market place.
The call for ‘transparency’, presents a comparable problem
where the issue may arise from over interpreted consumer utter-
ances rather than from a deeply needed wish for transparency by
consumers (see Dijksterhuis, 2004).
Many workers in marketing functions believe in these terms,
whilst many scientists with a psychological background agree that
these are just words consumers use when asked about food. Words
consumers do not use, or other matters they do not, or even cannot,
explicate, can be just as powerful in increasing product purchase.
The focus on the terms that arise from marketing studies, not al-
ways scientifically sound, is limiting the range of potentially suc-
cessful product innovations or product communication.
248 G. Dijksterhuis / Trends in Food Science & Technology 50 (2016) 243e248
8. Discussion and conclusion
As the main research problem behind the high fail rate of new
products lies in a lack of understanding of consumer motivation
and consumer choice, clearly behavioural science is where it's at, or
should be at. These sciences should therefore be more effectively
used to tackle this problem. In the context of food product devel-
opment there is the much debated increase in diet related ailments
(obesity, diabetes, heart disease). These are not results of the food
products themselves, but of the behaviour of consumers towards
these products. The solution to such problems lies in understanding
the determinants of consumer behaviour. Health is increasingly
recognised to not be a matter of food products themselves, but of
choice. Development of ‘healthy’ food products is not enough,
development of strategies to guide, or ‘nudge’, consumer choice is
needed. We regret to again have to conclude that there seems to be
an imbalance between funds available for research in food product
production compared to those for food behaviour studies (cf.
Dijksterhuis, 2014b).
The five areas we presented in this paper: (1) ‘pillars’ (too many
different functions addressing different aspects of the consumers
and of product development), (2) ‘higher management focus’ (not
geared towards understanding consumer behaviour), (3) ‘popular
science books’ (out-dated research directives resulting from a hi-
erarchical management model), (4) ‘quality and Quality’ (a defini-
tion of ‘quality’ that leads to invalid quality parameters), and (5)
‘psychophobia’ (the latent fear of trusting behavioural science re-
sults), may help understanding the mismatch between the pro-
ducers' expectations of new products, and the limited commercial
success of 50%e75% of them.
We can here only present a few suggestions for each of the five
areas. We do not claim that these are the only possibilities, nor that
they are always the right solutions. More detailed programmes
could of course be developed, but a more specific content of those
will depend on the type of product and industry concerned:
1. ‘pillars’: Integrate all functions that address behavioural aspects
of consumers (consumer research, marketing, sales, customer,
retail, brand, delivery, etc.) into one vertically integrated section
to optimise communication between these areas, and to ensure
maximal use of the most product-relevant and latest behav-
ioural science insights in all functions,
2. ‘higher management focus’: recognition of the differences be-
tween views and focus of technological, marketing, and
behavioural science groups; use internal experts to inform
higher management (not the other way around),
3. ‘popular science books’: less stress on (hidden) hierarchy,
appoint (behavioural) scientists in higher management of an
integrated behavioural science section (point 1 above) to ensure
the relevance of the directives for behavioural research, pre-
venting the latest consumer or marketing fad being followed,
4. ‘quality and Quality’: continually gauge (lab) quality parameters,
and validate them based on the success of marketed products,
although it is recognised that in many cases the information is
often not organisation wide available, which calls for steps to
assure that first,
5. ‘psychophobia’: Employ behavioural scientists and trust their
science as if it were any other (hard) science (indeed behavioural
science is often the harder science).
The author has requested enhancement of the downloaded file. All in-text references underlined in blue are linked to publications on ResearchGate.
When our conclusions are taken seriously, some corporate
research departments may have to take drastic measures. We hope
that our view on the matter can help raise the success rate of new
products, and lead to much more efficient investments in consumer
and marketing relevant R&D.
Conflicts of interest
The author declares there are no conflicts of interest.
Acknowledgements
I am very grateful for the discussions I had with a range of sci-
entists, both academic and corporate, over the years. In particular I
thank some anonymous colleagues who discussed the topic of this
paper with me in detail, and contributed valuable insights that I
integrated, with their consent, into the paper.
References
APQC. (2003). Improving new product development performance and practices.
Houston, TX: American Productivity & Quality Center. www.apqc.org/pubs/
NPD2003.
Bargh, J. (2006). What have we been priming all those years? On the development,
mechanisms, and ecology of nonconscious social behaviour. European Journal of
Social Psychology, 36, 147e168.
Brown, B., & Anthony, S. (2011). How P&G tripled its innovation success rate. Harvard
Business Review (June).
Chesbrough, H. (2005). Open innovation: the new imperative for creating and profiting
from technology. Harvard Business Review Press.
Cheung, T., Junghans, A. F., Dijksterhuis, G. B., Kroese, F., Johansson, P., Hall, L., et al.
(2015). Consumer's choice-Blindness to ingredient information. Appetite. http://
dx.doi.org/10.1016/j.appet.2015.09.022.
De Ridder, D. T. D. (2015). Health psychology. In J. D. Wright (Ed.), International
encyclopedia of the social and behavioral sciences (2nd ed., Vol 10, pp. 674e678).
Oxford: Elsevier.
Dijksterhuis, G. B. (2004). Transparency: who cares? A user manual for a piece of
chocolate. In G. J. Hofstede, H. Schepers, L. Spaans-Dijkstra, J. Trienekens, &
A. Beulens (Eds.), Hide or Confide? The dilemma of transparency. 's Gravenhage:
reed business information (pp. 159e164).
Dijksterhuis, G. B. (2012). The total product experience and the position of the
sensory and consumer sciences: more than meets the tongue. New Food, 15(1),
38e41.
Dijksterhuis, G. B. (2014a). Eating, food and experience in 2034. LinkedIn. July 15.
Dijksterhuis, G. B. (2014b). Van voedsel naar eten, van productie naar gedrag. (From
food to meals, from production to behaviour. In Dutch). LinkedIn. October 16.
Dijksterhuis, G. B. (2016). Multisensory flavour priming. In C. Spence, & B. Piqueras-
Fiszman (Eds.), Multisensory flavor perception: from fundamental neuroscience
through to the marketplace. Elsevier Publishers.
Kahneman, D. (2011). Thinking fast and slow. New York: Farrar, Straus and Giroux.
Kahneman, D., & Tversky, A. (1979). Prospect theory: an analysis of decision under
risk. Econometrica, 47(2), 263e292.
Kahneman, D., & Tversky, A. (2000). Choice, values, and frames. New York: University
Press.
€ster, E. P. (2003). The psychology of food choice: some often encountered fal-
Ko
lacies. Food Quality and Preference, 14, 359e373.
Nielsen. (2014). Breakthrough innovation report (European edition). september 2014.
http://www.nielsen.com/content/dam/nielsenglobal/eu/nielseninsights/pdfs/
Breakthrough_Innovation_Report_EU_FINAL.pdf.
Schacter, D. L. (1987). Implicit memory: history and current status. Journal of
Experimental Psychology: Learning, Memory, and Cognition, 13(3), 501e518.
Thaler, R. H., & Sunstein, C. R. (2009). Nudge. Improving decisions about health, wealth
and happiness. London: Penguin Books.
Woods, A. T., Lloyd, D. M., Künzel, J., Poliakoff, E., Dijksterhuis, G. B., & Thomas, A.
(2011). Expected taste intensity affects response to sweet drinks in primary
taste cortex. Neuroreport, 22(8), 365e369.