15 Apr Identifying the ingredients to predicting NPD success
Identifying the ingredients to predicting NPD success: A holistic view of the customer
In a nutshell
I’m on a journey to see what tools and theories are available that will help us to better predict success of concepts during the NPD process. After thoroughly reviewing the academic literature about what really works, we have created a range of new tools to better predict likely NPD success which we are currently validating. I’d like to share what I have learned so far. Over the next few posts, I will outline key observations I’ve made. I hope people will respond with observations/criticisms of the opinions and suggestions. On that note….
Purchase Intention is the measure that most companies still rely on when evaluating the likely potential of new product concepts, despite overwhelming evidence that stated intentions are poor predictors of future behaviours. As often stated, the bulk of a person’s decision making is often not conscious and this at least in part explains the poor performance of purchase intention as a predictive measure.
New ideas and techniques are coming onstream every day, often without being rigorously and independently validated. These new approaches all offer part of the solution: Purchase intention, System 1 vs System 2 Thinking, Emotional Engagement, Personality/Motivational Drivers and Relationship Theory are frameworks for understanding behaviours that may add to our ability to predict future success while neuroscience has provided tools to bypass rational responses.
This discussion sets out an argument to move beyond top 2 box purchase intention as a predictive measure and identifies areas worthy of exploration and validation.
Purchase Intention is the measure that most companies still rely on when evaluating the likely potential of new product concepts, despite overwhelming evidence that stated intentions are poor predictors of future behaviours. As often stated, the bulk of a person’s decision making is often not conscious and this at least in part explains the poor performance of purchase intention as a predictive measure.
New ideas and techniques are coming onstream every day, often without being rigorously and independently validated. These new approaches all offer part of the solution: Purchase intention, System 1 vs System 2 Thinking, Emotional Engagement, Personality/Motivational Drivers and Relationship Theory are frameworks for understanding behaviours that may add to our ability to predict future success while neuroscience has provided tools to bypass rational responses.
This discussion sets out an argument to move beyond top 2 box purchase intention as a predictive measure and identifies areas worthy of exploration and validation.
Many of research’s greatest breakthroughs have come from outside our industry
I have been a market researcher all my professional life, and part of my academic life too. Guess you would call me a “lifer”. I like that no two days are the same, that I get to work with some fantastic brands and get to try to understand why and how people do what they do. I have also been fortunate to have met and worked with some of the truly great minds of this industry and been influenced by the way they think (I think especially here of Hendrik Hendrickx & Jan Callebaut of Censydiam and Jannie Hofmeyr of Conversion Model fame). These people took learnings from outside of everyday Market Research and revolutionised the way we think about consumer behaviour.
Similar revolutions are always at play in Market Research. At the moment areas of revolution appear to be in the form of neuroscience, artificial intelligence and the concept of System 1 vs System 2 thinking, among many others.
Similar revolutions are always at play in Market Research. At the moment areas of revolution appear to be in the form of neuroscience, artificial intelligence and the concept of System 1 vs System 2 thinking, among many others.
We risk grabbing the latest shiny tool without validation or seeing if it adds to a holistic point of view of consumers – often at the expense of effective incumbent approaches
To be seen as forward thinking, many people have adopted new tools and techniques largely without solid academic or systematic validation. And in the rush for the newest tools and to be seen as different, some of the old ways may have been prematurely discarded. Ironically, some of the old tools and measures that have been proven to be less than reliable, live on to this day.
Most new approaches / theories are presented in “silos” based on the dogma that a single theory or data collection approach is supreme, rather than striving for a holistic view of the consumer, which is not only tolerant of different underlying philosophies, but accepting that maybe the whole is much more than just the sum of the parts.
Most new approaches / theories are presented in “silos” based on the dogma that a single theory or data collection approach is supreme, rather than striving for a holistic view of the consumer, which is not only tolerant of different underlying philosophies, but accepting that maybe the whole is much more than just the sum of the parts.
My goal: Shine a light on what is proven and look to uncover how all these market revolutions can improve the prediction of success of new product concepts
My real area of interest in the research world is NPD and specifically the prediction of success of ideas early in the NPD process. I also believe that at the early stages of NPD we should be looking to identify if and how we can optimise concepts to become commercially viable before we kick the ideas to touch. And I see many opportunities to embrace the new ideas without completely throwing out the old.
To ensure that I rigorously evaluate and develop better approaches to NPD evaluation, I am undertaking university study that will require me to delve deeply into the academic literature and hopefully shed light on the latest academic thinking and how it relates to both the old and new ideas that dominate the industry. In essence, what does the science really tell us helps predict future success.
Every now and then I will post some summaries of what I have been reading and researching and would welcome anyone to comment on what I have written down – whether you agree or disagree – as I am sure there are no black and white answers, just points of view.
To ensure that I rigorously evaluate and develop better approaches to NPD evaluation, I am undertaking university study that will require me to delve deeply into the academic literature and hopefully shed light on the latest academic thinking and how it relates to both the old and new ideas that dominate the industry. In essence, what does the science really tell us helps predict future success.
Every now and then I will post some summaries of what I have been reading and researching and would welcome anyone to comment on what I have written down – whether you agree or disagree – as I am sure there are no black and white answers, just points of view.
The Stage-gate process: Gold standard in NPD with Market Research at its heart
New Product Development (NPD) is the process of identifying a consumer need, creating a solution to that need in the form of an idea for a product or service, and then creating a tangible representation of the idea to take to market. For most companies, NPD is seen as necessary to create or maintain a competitive advantage. At the very least, NPD allows companies to remain relevant to their customers in a rapidly changing commercial landscape.
Most major FMCG companies in developed markets have a detailed system for NPD, many of which are adaptations of the Stage-Gate® process as originally outlined by Cooper (1988, 1990). Cooper et al (2017) among others have taken the idea of a sequential NPD process and developed the idea further. In his 2019 book, Cooper illustrates the Stage-Gate® process (see image alongside right):
The essence of most stage gate processes is therefore that progress is reviewed or researched at every identified “stage” before companies commit to investment in the next stage of the NPD process. This evaluation often involves research with an end user or user group i.e. the customer. By implication therefore, market research plays a critical role in the NPD process of many companies – and for most mid to large FMCG companies, this research is supplied by commercial market research companies.
Most major FMCG companies in developed markets have a detailed system for NPD, many of which are adaptations of the Stage-Gate® process as originally outlined by Cooper (1988, 1990). Cooper et al (2017) among others have taken the idea of a sequential NPD process and developed the idea further. In his 2019 book, Cooper illustrates the Stage-Gate® process (see image alongside right):
The essence of most stage gate processes is therefore that progress is reviewed or researched at every identified “stage” before companies commit to investment in the next stage of the NPD process. This evaluation often involves research with an end user or user group i.e. the customer. By implication therefore, market research plays a critical role in the NPD process of many companies – and for most mid to large FMCG companies, this research is supplied by commercial market research companies.
Risks of failure at launch are numerous, costly and sometimes intangible. But the earlier we can weed out the bad ideas the better
There are at least two key reasons that a sequential stage gate process is embraced by business. The first one being that ideas that make it through a stage gate process have a greater chance of being successful and secondly the stage gate process rejects more ideas, earlier in the process, which saves significant amounts of investment both direct and indirect.
There has been plenty of research into the costs associated with new product failure. As mentioned above there is naturally the direct investment cost, but other less front and centre costs include:
In a case study, Olthuis (1997) illustrates how much more efficient it is to reject ideas early in the NPD process (see table alongside).
It is not only the overall potential for success that needs to be considered but also speed. It is reported that some trade outlets in Australia review performance of new products after just three months! That is very little time to allow for distribution, awareness, trial and repeat behaviour to be established. New products need to hit the ground running else they won’t have the necessary momentum to outperform the products they are replacing in the times afforded to them by the trade.
The earlier in the process failures can be weeded out and successes concentrated for optimisation, the stronger the momentum a new product is likely to have at launch.
There has been plenty of research into the costs associated with new product failure. As mentioned above there is naturally the direct investment cost, but other less front and centre costs include:
- Opportunity cost – the money lost by not launching a successful product;
- Time cost – executive and production staff dedicated to pursuing a losing cause;
- Reputational cost – the trade lose confidence in suppliers if they launch poor products; and
- Internal confidence – lack of faith generated internally of the process.
In a case study, Olthuis (1997) illustrates how much more efficient it is to reject ideas early in the NPD process (see table alongside).
It is not only the overall potential for success that needs to be considered but also speed. It is reported that some trade outlets in Australia review performance of new products after just three months! That is very little time to allow for distribution, awareness, trial and repeat behaviour to be established. New products need to hit the ground running else they won’t have the necessary momentum to outperform the products they are replacing in the times afforded to them by the trade.
The earlier in the process failures can be weeded out and successes concentrated for optimisation, the stronger the momentum a new product is likely to have at launch.
Despite extensive testing the supermarket aisles are still “littered with the corpses of failed new product launches”
Despite widespread acceptance that NPD is critical to business success and the implementation of structured NPD processes to improve success, it is recognised that a very high percentage of products that are launched are destined to fail. There is a wide range of numbers presented as the real “rate of NPD failure”. Some of the most regularly quoted rates of NPD failure are as high as 80% or 90%. Harvard Business School Professors Zaltman and Christensen are often quoted as claiming failure rates of this magnitude e.g. “…despite enormous amounts of time and money dedicated to customer surveys and marketing, approximately 80% of all new products fail within six months or fall significantly short of their profit forecast” (Zaltman, 2003).
Others argue that the regularly quoted rates of failure often refer to the proportion of all “ideas” that fail, rather than the proportion of “launched products” that fail. With actual failure rates being closer 30% to 50% with FMCG failure rates being significantly higher than those of more durable products.
While there is no consensus on the rate of failure of new products post launch, there is wide acceptance that the failure rate is too high. Both the literature and personal interactions with Marketing Mangers still say that accurately predicting success of potential new products is still one of the things that keeps them awake at night.
Others argue that the regularly quoted rates of failure often refer to the proportion of all “ideas” that fail, rather than the proportion of “launched products” that fail. With actual failure rates being closer 30% to 50% with FMCG failure rates being significantly higher than those of more durable products.
While there is no consensus on the rate of failure of new products post launch, there is wide acceptance that the failure rate is too high. Both the literature and personal interactions with Marketing Mangers still say that accurately predicting success of potential new products is still one of the things that keeps them awake at night.
Beyond Purchase Intent
In the early twentieth century Sigmund Freud started to espouse the role of the unconscious mind on decision making and the fact that the largest piece of decision making is not conscious. In 1915 in his book “General Psychological Theory: Papers on Meta-psychology”, Freud first uses the metaphor of the iceberg to illustrate the portion of our thought processes that are conscious (above the water) and the much larger piece that lies beneath.
In his 2003 book “How Customers Think: Essential Insights into the Mind of the Market”, Zaltman makes the claim that “Ninety-five percent of thought, emotion and learning, occur in the unconscious mind – that is, without our awareness”. There are a plethora of papers describing unconscious factors that drive decision making, some of these include:
While not an exhaustive list, the above examples show that for over a century, researchers have known about, theorised about and investigated the fact that many decisions consumers make, including the choice to use a new product that is launched into the market, are not conscious or even directly about the product itself.
In his 2003 book “How Customers Think: Essential Insights into the Mind of the Market”, Zaltman makes the claim that “Ninety-five percent of thought, emotion and learning, occur in the unconscious mind – that is, without our awareness”. There are a plethora of papers describing unconscious factors that drive decision making, some of these include:
- Personal factors: Demographics & emotions
- Psychological Factors: Motivation & personality
- Situational factors: Competitive set, pricing, distribution etc
While not an exhaustive list, the above examples show that for over a century, researchers have known about, theorised about and investigated the fact that many decisions consumers make, including the choice to use a new product that is launched into the market, are not conscious or even directly about the product itself.
A long history of using Purchase Intent as a predictor of likely behaviour
As illustrated above, the majority of mid to large FMCG companies use commercial market research companies to provide customer feedback into the NPD process and in particular to try and predict likely success early in the NPD process. If the above decision-making observations are contrasted with the practices of commercial researchers, there is a disconnect.
It is estimated that up to 90% of commercial market research companies’ clients rely to some degree on the measure of “purchase intention”, or “purchase interest” to determine likely purchase behaviour in the future. This thinking has been long established with Morwitz, V. et al (2007) quoting Fishbein & Ajzen (1975): “If one wants to know whether or not an individual will perform a given behavior, the simplest and probably the most efficient thing one can do is to ask the individual whether he intends to perform that behavior.”
It is estimated that up to 90% of commercial market research companies’ clients rely to some degree on the measure of “purchase intention”, or “purchase interest” to determine likely purchase behaviour in the future. This thinking has been long established with Morwitz, V. et al (2007) quoting Fishbein & Ajzen (1975): “If one wants to know whether or not an individual will perform a given behavior, the simplest and probably the most efficient thing one can do is to ask the individual whether he intends to perform that behavior.”
Purchase Intent has a positive but weak correlation to future behaviour
David Ogilvy is famous (among other things), for saying:
“People don’t think how they feel, they don’t say what they think, and they don’t do what they say”,
……and many researchers have noted that purchase intention is not a reliable indicator of future behaviour. While the overall correlation between purchase intention and behaviour is a positive one, the strength of that relationship is often weak. Sheppard, B. H. et al (1988) reviewed previously published research to establish the overall average correlation across different product categories between purchase intention and future behaviour to be 0.52, but the range across the studies compared was 0.15 and 0.92.
The evidence also suggests that FMCG (lower involvement) categories and concepts that are more innovative/different (harder to visualise) have significantly lower levels of correlation than durable, high involvement and familiar categories of purchase (Morwitz et al, 2007).
The evidence also suggests that FMCG (lower involvement) categories and concepts that are more innovative/different (harder to visualise) have significantly lower levels of correlation than durable, high involvement and familiar categories of purchase (Morwitz et al, 2007).
The market research industry as a whole even questions Purchase Intent as a key predictor – yet its dominance remains
Purchase intention is so inconsistent and unreliable as a predictor of future behaviour, that many articles have been written about both why purchase intention is not a reliable predictor of future behaviour and how to “adjust” or weight down purchase intention results to be more realistic (Jamieson & Base 1989). In many western cultures there is a need to down weight purchase intention scores due to respondents’ propensity to overestimate their likelihood to buy a product if launched. Even the market research industry that uses this measure consistently, is starting to question its use. I was at conference in Australia a few years back (2014) and one of the guest speakers was a thought leader in the market research industry, Joel Rubison, and he said of purchase intent (I paraphrase here):
“It concerns me that, as an industry, we still rely on a measure where we discount 70 or 80 percent of response, to get anywhere near an accurate figure”.
Why do we still depend so heavily on Purchase Intent?
The concern, therefore, is that despite serious concerns about the effectiveness of the purchase intention measure and significant evidence that much of a consumers decision making occurs outside of a conscious choice about a product, purchase intention is still the primary measure used in predicting likely market success of new products by many marketers and researchers.
There are many reasons for this, some include:
There are many reasons for this, some include:
- Inertia – it is what has always been done (which means most normative databases have one common question included in them – purchase intent
- Simplicity – like NPS it is one number to hang your hat on – it is easy to set evaluation targets against, it is easy to explain to the exec and accountant!
- Absence of validated alternatives
Future posts
There have been many attempts to drive the discussion forward and I believe most add positively to that discussion and the discipline of success prediction. My goal moving forward is therefore to explore several topic areas that could contribute to NPD evaluation and suggest how I see the pro’s and con’s of these ideas. There are many executional reasons that products fail at launch, but from a prediction point of view I will be focussing on factors more directly related to the consumer.
Specific areas I intend to write more about in the coming months include:
Specific areas I intend to write more about in the coming months include:
- System 1 vs System 2 processing: This has been adopted by many due to its simplicity in much the same way that NPS has been adopted. But most evidence suggests Sys1 / Sys2 is a continuum rather than dual processing system. Even the person who introduced the idea of System1/System 2 (Keith Stanovich) says he wishes he had not used those terms as they have been taken too literally.
- Emotion: Specifically, our use of neuroscience in measuring emotion and the relationship this has in the adoption of new products.
- VR/AR: As a technology to bypass rational responses and tap into decision making in its most basic form
- Personality/Motivational research: While separate strands of thinking, they are related.
- Relationship theory: The study of understanding the relationships people have with brands.
An invitation
Any discussion is a good discussion, so any thoughts are welcome – positive or negative. The more concerns that discuss from both research and marketer side the stronger the tools we develop moving forward.
paul@imagineresearch.co.nz
paul@imagineresearch.co.nz
Interesting reading
Castellion, G. & Markham, S.K, (2013): “Perspective: New Product Failure Rates: Influence of Argumentum ad Populum and Self‐Interest”, The Journal of Product Innovation Management 30 (5)
Cooper, R. G. (1988): The New Product Process: A Discussion Guide for Management, Journal of Marketing Management, 3 (2) pp 238-255
Cooper, R. G. (1990): “Stage-gate systems: a new tool for managing new products”. Business horizons, 33(3), pp 44-54.
Cooper, R. G. (2017): “Winning at new products: Creating value through innovation” (5th ed) New York, NY: Basic Books. Cooper, R. G. (2017): “The drivers of success in New Product Development”, Industrial Marketing Management 76, pp 36-47
Cooper, R. G. (2019): “The drivers of success in New Product Development”, Industrial Marketing Management 76, pp 36-47
Ekman, P. Friesen, W.V. & Ancoli, S. (1980): “Facial Signs Of Emotional Experience”. Journal of Personality and Social Psychology. Soc. Psychol. 39 pp 1125–1134
Fishbein, M., & Ajzen, I. (1975): “Belief, Attitude, Intention, and Behavior”. Reading, MA: Addison–Wesley
Freud, S (1915): “General Phycological Theory: Papers on Meta-psychology” pp 116-150 Collier Books
Haddad, L., Hamza, K. M., Xara-Brasil, D. (2015): “Archetypes and brand image: an international comparison”. Australian Journal of Basic and Applied Sciences, 9(34) pp 22-31
Kahneman, D. (2011): “Thinking, Fast and Slow”, Penguin
Krockow, E (2018): “How many decisions do we make every day?”, Psychology Today
Lewis, M. A. (2001): “Success, failure and organisational competence: A case study of the new product development process”, Journal of Engineering and Technology Management -JET-M,18(2), 185–206
Mark, M. and Pearson, C. (2001): “The Hero and the Outlaw: Building Extraordinary Brands Through the Power of Archetypes”, McGraw-Hill
Morwitz, V.G. Steckel, J.H. Gupta, A. (2007): “When do purchase intentions predict sales?” International Journal of Forecasting 23, pp 347–364
Olthuis, G (1997): Production Creation at Phillips Electronics. R&D Management, 27 (3). pp 213-224
Sheppard, B. H., Hartwick, J., & Warshaw, P. R. (1988): “The theory of reasoned action: A meta-analysis of past research with recommendations for modifications and future research.” Journal of Consumer Research, 15, pp 325−343
Urbig, D, Bürger, R. Patzelt, H. & Schweizer, L. (2013): “Investor Reactions to New Product Development Failures: The Moderating Role of Product Development Stage”, Journal of Management, 39 (4), pp 985-1015
Zaltman, G. (2003): “How Customers Think: Essential Insights into the Mind of the Market” Harvard Business School Pres
Cooper, R. G. (1988): The New Product Process: A Discussion Guide for Management, Journal of Marketing Management, 3 (2) pp 238-255
Cooper, R. G. (1990): “Stage-gate systems: a new tool for managing new products”. Business horizons, 33(3), pp 44-54.
Cooper, R. G. (2017): “Winning at new products: Creating value through innovation” (5th ed) New York, NY: Basic Books. Cooper, R. G. (2017): “The drivers of success in New Product Development”, Industrial Marketing Management 76, pp 36-47
Cooper, R. G. (2019): “The drivers of success in New Product Development”, Industrial Marketing Management 76, pp 36-47
Ekman, P. Friesen, W.V. & Ancoli, S. (1980): “Facial Signs Of Emotional Experience”. Journal of Personality and Social Psychology. Soc. Psychol. 39 pp 1125–1134
Fishbein, M., & Ajzen, I. (1975): “Belief, Attitude, Intention, and Behavior”. Reading, MA: Addison–Wesley
Freud, S (1915): “General Phycological Theory: Papers on Meta-psychology” pp 116-150 Collier Books
Haddad, L., Hamza, K. M., Xara-Brasil, D. (2015): “Archetypes and brand image: an international comparison”. Australian Journal of Basic and Applied Sciences, 9(34) pp 22-31
Kahneman, D. (2011): “Thinking, Fast and Slow”, Penguin
Krockow, E (2018): “How many decisions do we make every day?”, Psychology Today
Lewis, M. A. (2001): “Success, failure and organisational competence: A case study of the new product development process”, Journal of Engineering and Technology Management -JET-M,18(2), 185–206
Mark, M. and Pearson, C. (2001): “The Hero and the Outlaw: Building Extraordinary Brands Through the Power of Archetypes”, McGraw-Hill
Morwitz, V.G. Steckel, J.H. Gupta, A. (2007): “When do purchase intentions predict sales?” International Journal of Forecasting 23, pp 347–364
Olthuis, G (1997): Production Creation at Phillips Electronics. R&D Management, 27 (3). pp 213-224
Sheppard, B. H., Hartwick, J., & Warshaw, P. R. (1988): “The theory of reasoned action: A meta-analysis of past research with recommendations for modifications and future research.” Journal of Consumer Research, 15, pp 325−343
Urbig, D, Bürger, R. Patzelt, H. & Schweizer, L. (2013): “Investor Reactions to New Product Development Failures: The Moderating Role of Product Development Stage”, Journal of Management, 39 (4), pp 985-1015
Zaltman, G. (2003): “How Customers Think: Essential Insights into the Mind of the Market” Harvard Business School Pres
