For decades, marketers have relied on the Intent to Buy rating scale to assess customer purchase behavior when presented with a product or in-store concept. The Intent to Buy question is just one part of the traditional survey used by researchers, and focuses on what percentage of people say they would likely purchase a product. For decades it has been the best metric given the tools available. But that’s no longer the case—virtual research can now provide companies with observed purchase behavior.
In a study with InContext, American Fare Popcorn leveraged virtual reality to find that there was a significant difference between those who said they would purchase the popcorn (45%) and actual penetration (18%). Yet purchase intent is still consistently serving as the primary way marketers are making their in-store decisions. Below, we share the problems with relying on purchase intent, and why virtual research sales data is a better tool for understanding future shopper behavior.
Why Purchase Intent Isn’t Enough
The biggest issue with Intent to Buy ratings is they are attitudinal in nature. We ask the respondent to rate the likelihood of them buying a particular product within a certain time frame, which then should give us insight into whether or not they will actually purchase it in the future. However, all evidence suggests this attitudinal data is not strongly correlated with actual behavior—actions speak louder than words, if you will. There are many reasons for this:
-
-
- People don’t accurately recall what they have done in the past.
-
- People are notoriously bad at predicting what will happen in the future.
-
- Purchase decisions are often influenced by factors not included in the Intent To Buy question.
-
- People tend to be biased on ratings scales to please the interviewer.
- The Intent to Buy question is only about penetration and not about Buy Rate.
-
Let’s take a deeper dive into how each of these factors can lead to inaccurate data.
People don’t accurately recall what they have done in the past.
One often-studied aspect of psychology is memory. What people remember about past events is likely to change down the road—like a game of telephone, details change slightly until the person believes the memory they have is what actually happened. In that same vein, people answer survey questions poorly when it comes to past purchase behavior. There’s a well-known industry rule of thumb: people’s recollection of when they last purchased a category is usually half as long ago as it actually was. And they struggle to remember what brands they purchased unless they have some visual cues, like a thumbnail of the package. This can result in false data.
People are notoriously bad at predicting what will happen in the future.
Simply put, people are not clairvoyant. Just as they tend to have false memories, it is also very difficult to predict what will happen in the future. From weather forecasts to sports championships to the stock market, those with expertise in the area still struggle to predict an outcome. The general public is even less likely to be able to see beyond the present. When it comes to purchase intent, “the fundamental problem is that, in spite of what conventional wisdom tells us, it is not the voice of the consumer that matters. What matters is the mind of the consumer.”1 So how can we move beyond Intent to Buy, which by definition is only a prediction of what they will do in the future?
Purchase decisions are often influenced by factors not included in the Intent to Buy question.
When respondents rate their intent to buy, they are often not thinking about the way they shop. They have usually made a list, know their favorite brands and don’t want to spend any more time than necessary making a purchase in any one category. Then there are situational influences, like their reaction to physical and social surroundings; the time when the purchase decision is being made; the reason for making a particular purchase; and the buyer’s mood at the time. Often shoppers have kids to contend with, or are on time constraints. These factors all affect the likelihood that a new product will get noticed on the shelf, and are often missing when it comes to figuring out Intent to Buy on a survey.
People tend to be biased on ratings scales to please the interviewer.
When respondents are asked to rate anything, they tend to want to give a positive rating to please the interviewer or company doing the research. This is sometimes referred to as the Hawthorne Effect, in which subjects change their behavior when they know they are being studied. This means more people will give a higher rating (a 4 or 5 on the 5-point scale) than they really should. It isn’t that they don’t believe the answer they are giving at the time, but they subconsciously are willing to give a more positive rating than they are likely to follow through with in a store.
Intent to Buy is only about penetration and not about Buy Rate.
Perhaps the biggest problem with Intent to Buy is that it only measures penetration. It is designed to be an attitudinal-based estimate of what percentage of people will buy the product in the future. Unfortunately, the sales of a product can be impacted not only by penetration, but also by Buy Rate— the average amount of that product purchased over a whole time period. Buy Rate helps grow your business by getting your current buyers to buy more than they did before.
So how can we move beyond Intent to Buy, which by definition is only a prediction of what they will do in the future?
Reality in Virtual
Luckily, there’s a solution to the challenges presented when relying solely on purchase intent. At InContext, our virtual reality shopping simulations make it possible to observe purchase behavior, through both attitudinal and behavioral data. When online consumers “shop” in our virtual store environments, they don’t know how to adjust their shopping to please the interviewer—so they just behave as they would in a real store. (The correlation between virtual and reality is over .90.) Because of this, there is no survey bias and no false memories—we can ask attitudinal questions after every exercise. This results in unbiased behavioral data. And our virtual sales data accurately measures not only penetration changes, but also Buy Rate changes. Purchase history is still a relevant data point when it comes to testing new products or concepts, but virtual simulations dramatically enhance that data by also looking at the present, and into the future.
