Did you know there is a 0.18 correlation between purchase intent and purchase*? Surprisingly, this is a finding that has been known for over 30 years.
Since those 30 years have passed, subsequent studies have proven this true, time and again.
When framed in Behavioural Science terms, it makes sense. Consumer decision-making is largely emotional and largely sub-conscious. It is well known that what consumers think they will do and what they actually do are often poles apart.
Many research approaches still work on the assumption that what consumers claim will accurately predict their likelihood to buy something. This continues to happen in surveys looking at a concept, idea or prototype in isolation to the rest of the market… and it simply doesn’t hold weight.
Where does the purchase intent gap leave us?
Despite all of this, claimed intent is probably the number 1 metric that brand builders will look to when deciding what to do next when developing innovation and creative concepts. There are global databases used day in and day out to show how concepts or ideas fare against all others that have been previously tested, using purchase intent as a key indicator of success versus the norm.
Unfortunately, it prevails because it is simple to understand. It provides a heuristic pathway to navigate what is otherwise a tricky and nuanced decision-making situation. So often it’s a flawed measure that leads to huge investment decisions, unlikely-to-succeed innovations and creative content that is then moved into production.
Yvan Goupil, Catalyx’s Head of Client Strategy, explores the flaws of PI
“Purchase intent is too blunt, it doesn’t take into account the context within which the purchase will ultimately be made. It highlights too much the focus on price – that being only one driver in the ultimate purchase act. Consumers can surprise themselves by going for something they may have found too expensive, but they may do so because it satisfies another element of their needs – there’s some other incentive to buy it. That could be emotional; the pride it gives you to own it, for example. There are also external influences which will distort the pure purchase intent path. Like the shopping environment surrounding it, or advocacy from someone you trust.
“You might not have considered a facial cream that costs more than your regular one, but it could pop up on a nice promotion and suddenly you’ll buy it, despite your original plan. Your intent would not have led you there, but you end up doing it.
“From a research standpoint, purchase intent and its many flaws are hard to translate into reality because there’s just too many elements in play. When questioning consumers we commonly see Likert Scales used – a five-point scale that ranges from a variation of “I will definitely do X” to “I will definitely not do X”. The top two boxes of this are “I will definitely” and “I will probably”, and that on its own is incredibly hard to measure between different people. There’s inconsistency between what terms like these mean to individuals – it’s very hard to truly gauge such a subjective claim.
“Purchase intent has value, but it shouldn’t be the only metric you use. For Catalyx, it’s not just the subsequent analysis that we handle PI with, it’s the attention to detail with the framing and contextualising of the question of PI in the first place. It needs to be handled from end-to-end with great precision to get you as close to consistent in consumer interpretation as possible.”
So, what should be the future of PI?
As a firm, Catalyx still uses the Purchase Intent question – but with great care, and we never use it in isolation. We make sure our clients don’t either. Catalyx always considers it with a range of other KPIs and we put as much, if not greater, weight on the qualitative outputs we receive in parallel and our strategic interpretation of insights into actions.
We continue to make future-focused strides that will put an increased weighting on the unstructured data inputs from our consumers. Their emotional reactions, their real life behaviours, their verbatims and lateral responses to what they see. As we build our Large Language Models we will be able to move away from the purchase intent claim completely. The future will be full of statistically robust unstructured data codes to give us a far more accurate view of how the consumer will react in the real world.
At Catalyx, many of our clients are already forward-looking – decisively (and bravely) letting go of historic PI norms-based decision-making to engage with a model that is far more behavioural and true to how consumers react in real life. The results they are seeing are significant. Just maybe, purchase intent will start to lose its flawed grip on decision-making. From there, brands are enabled to build innovations with the best chance of success.
*(Morwitz,V et al “The link between purchase intentions and purchase behavior: Predicting across individuals and over time”, 1991)
Go beyond PI to innovate with confidence
We’re ready to unlock your innovation’s success, and won’t settle until our work does just that.
Don’t just take our word for it; check out our new website’s Case Studies to discover how our qual at quant scale approach has supercharged the success of innovations across industries.
Follow us on LinkedIn to make sure you never miss an announcement.
