The Eyes Don’t Lie
Several CPG companies are adopting eye-tracking technologies as a way to overcome the unreliable answers that often come from focus groups. Pairing such devices with mockups of supermarkets is enabling brands to roll out new products faster and more cheaply while coming up with packaging and shelf layouts that draw attention.
According to an article in The Wall Street Journal, advances since the nineties now enable retinas to be tracked to gain a clear view of where consumers "are looking, for how long and how often." The collected information is formed into a "heat map" with colors indicating where people looked on a simulated shelf.
Using a retina-tracking camera, Kimberly-Clark determined which packaging designs on Viva paper towels attracted consumers’ attention during the first 10 seconds they viewed the shelf, according to the report.
In exploring a redesign of its Axe body wash, Unilever had testers wear specially equipped glasses that sensed sideways and vertical motion within a virtual 3D environment. Results led Unilever to change the canister’s shape from curvy to straight and also guided decisions around colors and font size. Eye-tracking was also used by Unilever to change shelf space for a deodorant.
"With a virtual shelf set, in a few seconds, with a click of the mouse, you can modify your product, your pack, your display, and really co-create it with the consumer almost in real time," Joanne Crudele, Unilever’s director of global skin consumer technical insight told the Journal.
Lower costs are also driving increased usage of simulation and eye-tracking. P&G told the Journal that most physical prototypes cost more than $1,500 and can run significantly higher. Now, 80 percent of P&G’s new products are developed using some form of software modeling or simulation.
But eye-tracking technologies are also seen as more reliable than focus groups, where participants often overestimate their interest.
"There’s often a big disconnect between what people want to do and what they say they want to do," Steve Posavac, a professor of marketing at Vanderbilt University, told the Journal. "Any attitude," he said, "becomes more extreme" in research studies.
Brain-wave monitoring and facial-recognition technologies are likewise increasingly used in consumer research over focus groups to more accurately gauge intent.
Writing for Modern DC Business, Marcia Moran, a consultant, said such simulation technologies might decrease the barrier to entry for new or smaller companies and speed up product development overall. But she wondered whether retina-tracking technologies track "all of the relevant inputs" that drive human behavior.
Discussion Questions
Discussion Questions: What do you see as the pros and cons of eye-tracking technologies in testing packaging and shelf sets? Overall, how do you think increased use of simulation techniques will affect the go-to-market cycle?
Simulation will definitely help the go-to-market timeline since you can incorporate customer feedback before any physical prototypes get built.
I’m not convinced that eye tracking technology is the silver bullet as the behavior during simulation may be different than behavior in front of the actual shelf.
People frequently look at other people and then their spouse or whatever catches them doing their optical evaluations and illusions. So eye-tracking isn’t really totally new for testing things.
Now eye-tracking is being scientifically applied in testing packaging and shelf sets, which is today’s latest testing paradigm. That’s good and it’s in tune with today’s marketing science.
As life goes on, increased use of simulation techniques will occur. They will likely affect the go-to-market cycle for a good while. Then a “better” program for that purpose will likely be created. Like life, idea development always goes on.
Perhaps the eyes don’t lie, but the problem is that they see too much…
Seriously, one major problem with eye-tracking is that it will lead to a lot of meaningless findings/correlations that will emerge naturally from having too much granular data. Researchers will stumble into “significant” effects that arise from chance variation alone, and will start to make decisions based on these non-generalizable observations.
It is more important than ever to develop real “data science” skills, which only come from being very familiar with the kinds of behavioral patterns (and underlying psychological theories) that have emerged from other (less data intensive) forms of retail research.
For starters, people interested in this area should read everything ever written by Herb Sorensen….
I have been using eye tracking technology in my research for over 30 years. As the technology has advanced, it’s allowed me to create innovative research approaches for clients…retailers, manufactures, and package designers. The research provides answers otherwise unavailable.
Make no mistake, eye tracking is a strong component of research that determines what attracts consumer attention and gains purchase consideration. However, you must still apply verbal question and answer to get at the significance of the attention.
Visual involvement identifies product and brand consideration, but without knowing how consumers interpret what they see, we don’t have the whole story. There are still issues of recognition, relevance and influence.
I use eye tracking to test effectiveness of in-store signage, displays as well as packaging. The technology can be stationary or mobile and applying and interpreting eye tracking properly requires experience and research know how.
We’ve show cased the difference between decisions that are impulse versus deliberate (think snacks and cough/cold remedies), we’ve looked at displays that facilitate purchase decisions and those that are totally ignored, and we’ve assisted marketers and designers identify packages that close sales at that point of sale as well as those that are turnoffs.
The advantage of eye tracking is that it documents the first crucial steps in the buying process…impact and consideration. However, it’s the question and answer in the research that closes the deal for pinpointing what generates brand and product understanding and conveys the message that resonates with shoppers.
Here’s a simple cheat sheet for use in research for packages and point of sale displays and signage. Be sure your research tests for: attention, impact, and identification of brand product, understanding of intended messages and imagery as well as influencing purchase consideration.
Beyond that packages and displays typically have specific goals…perhaps enhancing appetite appeal or convenience of use, looking more contemporary, etc. Properly created and conducted research evaluates that as well.
There’s still a place for focus groups and one on one in-depth interviews. You have to know when to use them and when they don’t apply. And don’t be afraid to ask the experts. That’s what we’re here for.
Why do it the simple way when the complex way works?
Apparently people in focus groups are deceptive at the least and outright lie at the most. So let’s keep using them, but make them wear tracking goggles (of course, just wearing them doesn’t influence the experience does it?) or attach electrodes to their skulls (at least that’s not invasive). Ah…welcome to advanced market research!
Following where the eyes track tells you only that that’s where the eyes track. That does not necessarily translate into the urge to buy. Heck put a picture of Halle Berry on the soap box and you don’t need no stinkin’ goggles to follow how most guy’s eyes track.
Seems to me we want to master energy transfer. Which design, color, placement, shape transfers positive strengthening energy to the viewer? Emphasis on the word ‘positive’ because negative energy attracts too, it’s just that one results in a sale and the other doesn’t.
The subconscious is where the ‘buy decider’ is located. To get at that we might try kinesiology…muscle testing in other words. I’ve used that to test if executive teams really believe what they claim are their “values.” I can tell you what’s revealed can be shocking. I’ve also used this to prioritize strategic alternatives. Energetic connection tests strong, disconnection tests weak. Every time.
So take your various prototypes and muscle test against them. If you really want to live in the ethers, put each one in an identical brown bag and have the person just hold it while you test for strength or weakness.
If you’d like a reference on this, one of the best is Dr. David Hawkins book, “Power VS Force: the hidden determinants of human behavior.”
Eye tracking technology can certainly provide detailed measures of the physical views of consumers such as where they look at a shelf set, font selections on packaging, etc., but focus groups will provide feedback about the actual product and not just the packaging. I think eye tracking technology combined with traditional focus groups will be the most effective technique for the go-to-market cycle.
Neither use of eye-tracking tells you whether you have a good package or a good shelf set. It tells you what people are attending to and you have to decide whether that is good or bad. In a controlled test, like the Axe example, it tells you that more people are looking at the package with one design vs another. Don’t confuse this with sales. There is no published data that shows that more attention leads to more sales beyond “if you don’t see it, you can’t buy it.” A lot of hype by eye-tracking companies — no validation with sales.
I agree with Dr. Fader that Herb Sorensen is a worthy source of understanding the nuances of eye tracking and other intensive research methodologies. I also believe that eye tracking, and even other less complex forms of observational research must be corroborated by measured, empirical shopper behavior modification once changes are made on the basis of research.
So many times research is used to offer rationale for making the change, but little is done to measure the impact of the change post hoc. In simpler, more concise terms, “did sales significantly increase after implementing changes to package design, shelf placement, etc?” If yes, and consistently yes, I would vote the the methodology is valid. If not, then perhaps nuances of the methodology need to be further explored.
I would take eye-tracking with a grain of salt. In 2-D environment such as paper or computer screens, we have been condition to how we track information from upper left to right while other cultures been conditioned to track from right to left.
However, in real world environments where depth comes into play, eye tracking science become questionable. As someone said, it doesn’t matter if an item is positioned where the eye start tracking first — the “free!” message is going to make them turn their focus/attention just as fast.
The use of 3-D simulation and quantitative measuring and tracking technologies will definitely affect product, packaging, placement and signage designs and implementations. The use of these technologies can dramatically increase the speed at which a product can be introduced while at the same time providing quantitative evidence to maximize its merchandising efficacy. By eliminating unreliable qualitative input along with the time and cost savings from prototyping products and analyzing focus groups’ insights, virtual quantitative evidence is not only cost-effective but could save months of development time.
These companies seem to think that people who look at products will also purchase them…have they made that correlation?
Eye tracking is just one biometric technique to employ at the retail level. Listening to our tone of speech and reading our body language are other methods to utilize. Also, this technology is not all that new, since it has been in development for more than fifteen years.
I do believe, as in the Tom Cruise movie, “Minority Report,” that biometrics have their place in retail. I think that “hot spots” on the sales floor can be better identified and leveraged to drive incremental sales. There are confidential pilots that have shown this to be true, and those retailers see this as a competitive advantage.
So the eyes don’t lie. What are we becoming when we now have to use retina technology to plan our marketing? Let’s complicate our life and of course our marketing.
I am reminded of the song the title, writer and singer I don’t recall; but some of the words go like this “In the year 2525, if man is still alive…” Maybe we will have thought monitors so when we want a corned beef sandwich on rye bread it will appear along with the iced tea…. Sarcasm? Maybe. You be the judge.
Professor Posavac’s comment is dead-on, but it doesn’t go far enough. Not only is there a disconnect between what people say they do and what they actually do, but there is a big difference between the virtual world and the real world. Eye-tracking may give you better hypotheses for packaging or shelf sets, but until you go out and test them in the real world, they are only hypotheses.
Ultimately, it isn’t where the consumer looks that matters. It isn’t even where she looks or where she lingers in real world stores that matters. It is what she actually buys in real world stores. Nothing else can stand in for that real-world impact.
I was disappointed to see the gratuitous pot shot about focus groups. For a vast range of issues, they are entirely reliable. They only become unreliable when you attempt to use them to learn things they can’t help you with, when you attempt to make them “semi-quantitative” (which happens far more often than it should), or when they are poorly constructed.
Having said that, what about eye tracking? There is a tremendous amount to be learned. What I’ve seen so far just confirms common sense — but often that’s quite important within a marketing department where the speed of ideas make common sense hard to grasp.
It is also important to know that these studies look at a single thing: eye tracking. That is one factor, but only one factor, in consumer choice. There are a vast number of other factors in play — factors that eye tracking won’t even hint at.
I expect it to keep being used in controlled circumstances and to find valid and useful learnings. And it’s not a silver bullet because there really isn’t one. Retail success will continue to require a tremendous arrangement of difficult to predict issues.
I appreciate the shout outs by Pete Fader and Mark Heckman. I independently studied vision biochemistry in the ’60s, including a seminar by George Wald, Nobel Laureate for his work on the subject. My own interest in the eye accelerated when my team invented PathTracker in 2001, which ultimately led to the tracking of millions of shopping trips on a second by second basis.
Translating that data into actionable insights for brands and retailers led to our process of imputing what is seen from the path the shopper takes, statistically computed as EyeShare. This patented metric is very useful for getting reasonably accurate metrics of the entire visual space in a store, and is a valuable adjunct for in-store media work.
Since then we have been accurately measuring both what is seen (the field of vision) AND what is focused on, sometimes described as attentional fixations, with both fixed and mobile eye-tracking equipment, of the type Joan Triestman describes. This extends into virtual reality methods.
What has become apparent is that visual control, and a whole lot more of the shopping trip, is NOT consciously managed by the shopper, but is under autonomic control. This, in turn, has led me to more in-depth study on the neurobiology behind the guidance/response mechanism, THE fundamental science of the purchase process.
And, by the way, both Peter Fader and Mark Heckman have been valuable collaborators with me on multiple occasions, including both providing chapters in my book, “Inside the Mind of the Shopper.“
What a major boost this will be for online delivery companies — I’m sure I’m not the only mega-neurotic paranoid privacy fanatic who will see eye-tracking as the last straw helping me decide never to set foot in a supermarket again.
Focus Group and Eye Tracking are both useful research tools, but they solve two different problems.
Human Beings simply are not able to use our rational brains to explain decisions that were (or will be) made by our subconscious. Asking a focus group which design for a tissue box they would choose in supermarket, simply doesn’t yield any useful results.
Because when that shopper actually makes it to the shopping aisle, that unconsidered decision to put a tissue box in her shopping cart is far more influenced by a variety of factors her rational mind is never aware of.
Focus groups generally aren’t good answering “what,” but they can give insight into the “why.” Why do you prefer this design to that design? Why do you buy this brand vs that brand. Their answers won’t be complete (or even necessarily accurate) but they may provide insight into some aspects of the product design that are useful.
By contrast, biometric data such as Eye Tracking gaze plots are very useful at telling us the what, but not the why. Which element of our design will get a shoppers attention first? What is the visual hierarchy of our messages?
I would also caution that Eye Tracking data from a lab is tricky to apply to a retail environment. Attention plots can change dramatically when you put the baby in the shopping cart, add retail lighting, inconsistent product stocking of adjacent products, etc….
The best retail eye tracking plots come from capture devices on real shoppers in real shopping environments. That will tell you what is getting a shoppers attention, but not why it is doing so.
There are a lot of excellent, insightful posts here, and I especially like the perspective in Jason Goldberg’s comment. First, observation of reality always trumps opinions and discussion by and of the observed. But as Jason notes, although careful observation and measurement can definitively establish the “what,” the verbal communication of focus groups or surveys can provide a huge addition to the “why.”
We are increasingly finding that the research design can be more important than the accuracy of the stimulus producing the data – very relevant, particularly to virtual reality research. (Thanks to my Ehrenberg-Bass colleague, Katherine Anderson. 😉
Further, on just the eye-tracking stuff. In general, the two broad measurement tools are fixed systems, such as the Tobii device. The alternate is something like the ASL MobilEye system, that is more suitable for “real world” tracking in the store. Without a detailed discussion of the relative merits, let me say that the fixed system is most relevant to shoppers’ experiences when they themselves are “fixed” in the store anyway. This would include things like evaluation of single packages, or small shelf-sets, where the shopper would not be moving around much in the real store, anyway.
However, use of a fixed system for anything more than 2-3 feet of shelf space (horizontal and vertical) is likely to produce bogus, imaginary, worthless results. This is because you cannot assess how a shopper will move, unless they have the opportunity to move. At least virtual reality provides them the opportunity for virtual movement, which MAY be adequate.
Nothing can beat the reality of real shoppers on real shopping trips in real stores, captured by relatively unobtrusive eye-ware. It’s good not only for all aspects of navigation, but also for the up close and personal relationship between the product and the shelf during the crucial final few seconds when the purchase process reaches its denouement. I’ve interrupted my writing of a major article for the Journal of Advertising Research on this exact topic, in order to do this post. 😉
The single greatest barrier to more widespread use of mobile eye tracking systems is the incredible difficulty of proper detailed analysis. With the fixed eye tracking systems, the field of view is fixed, defined and unchanging. Easy to produce those helpful heat maps, and a lot more. With any mobile system, the field of view is NOT fixed, and can change DRAMATICALLY as often as several times a second, and in any case, at least every few seconds. Until machine recognition of fields of view become efficient and practical, the final mile from the world to the brain (mediated by the eye,) will continue to be hampered by limited accurate information, some insightful speculation, and lots of poorly disguised ignorance. 😉
I’m working on it!
My doctoral dissertation in the mid-80s involved sending shoppers into 7-Eleven stores wearing football helmets with cameras mounted on top. As 7-Eleven’s national VP of advertising, I was testing the effectiveness of our in-store sign program, the largest in the world at the time and a huge expense for my department. I wanted to know if shoppers were looking at our signs at various levels throughout the store. I wanted to examine the question, “Why do stores hang signs from the ceiling?,” and its colloquial answer, “Because that’s the only place left.”
We learned a great deal and put it to good use. And to expand on Ralph Jacobson’s comment that “this technology is not all that new, since it has been in development for more than fifteen years,” it has really been in development for more than twenty-seven years if my research is taken into account.
And before that, you’ve got to think about the box design for Tide detergent. Here’s the story: Researchers determined early in the twentieth century that a baby’s attention was drawn first to its mother’s eyes, and that this attraction to the shape of an eye continued into adulthood. So, in the 1940s, P&G designed the Tide box to resemble an eye. You know the logo with its familiar circular design.
But, like others, I’m skeptical — not of the new technology, but of placing more reliance in it than deserved. I’m reminded of the little girl accompanying her mom in the store and tugging on her mother’s skirt whining “Mommy, Mommy, Mommy, I want this one.” It happens a lot and has nothing to do with Mommy’s eye tracking.