PureSpectrum - Schedule A Demo
Our new GreenBook Directory site is live!
COVID-19 guidance, tips, analysis - access full coverage here

Seeing is not Believing

In a study, Hotspex discovered that online testing is as effective as laser eye-tracking and, in the process, they uncovered a disruptive truth.

By Gera Nevolovich

Physical eye tracking at shelf, as measured by laser technology, has been the long-standing measure of success in packaging design research. A Hotspex study set out to determine whether our click-based online attention tracking tool is comparable to laser eye tracking. We discovered that online testing is as effective as laser and, in the process, we uncovered a disruptive truth.

But first, some key observations…

1. Online attention tracking and laser eye tracking results are comparable

Across the five categories included in the study, there is a good correlation (.67) between attention as measured by laser eye tracking and clicks, with some variability by category.  Both methodologies show that respondents usually begin their exploration by focusing on the center of the shelf. Their attention then gravitates outward to other areas of the shelf.

2. Online attention tracking is a better predictor of choice

This validation study included a measure of market share based on virtual shopping: ‘products purchased on shelf’. Online tracking is superior, correlating more closely than laser to products purchased on shelf in 4 out of the 5 categories, and is comparable with laser in the pasta sauce category.

The underlying reason is that the physical act of ‘seeing’ a package on shelf is fundamentally different from the act of ‘clicking’, which involves mental processing of stimuli. This second scenario (clicking) more closely imitates the real-life shopping experience, which involves the same level of mental processing, because it culminates in a choice.

The ‘seeing + thinking’ dynamic intrinsic to online attention tracking is even more important when consumers are faced with highly-cluttered shelves. While shelf clutter and diversification were the most prevalent in the gum category in this study, retail shelves in general are becoming busier, forcing consumers to spend more time ‘processing’ before making a choice. Laser tracks physical eye movements, but the heart and mind do not always follow the eye.

The relevance of these findings is underscored by the fact that our study observes a closer correlation between click data and purchased-on-shelf data than the correlation of laser to purchase data.

3. Disruptive truth: Online eye tracking is a better indicator of Market Share.

a) What are the implications for online packaging testing?

The purchase ‘sweet spot’ is where a package design achieves high scores on both claimed purchase intent as well as high rates of on-shelf purchase in the virtual shopping exercise. To get into this space, a package needs to be ‘noted’ (attention) and then be ‘processed’ (connection).

Our internal research and development studies suggest that consumers’ emotional relationships with a product account for at least 50% of consumer choice drivers. This underscores the importance of making a connection with the consumer at shelf in order to influence choice.

Online, click-based eye tracking is inclusive of both attention and connection, meaning that respondents are more inclined to click first on what they like as opposed to what they first see, and it is clear that what they like is what they tend to buy.

b) Online is on-the-money

Technology has leveled the playing field for packaging testing methodologies with interactive technology, graphical power and global scalability. Online packaging design tests can be conducted in multiple countries with hundreds of consumers per design, providing a much more comprehensive evaluation of a design’s performance on-shelf (shelf test) and in isolation (concept test) in a fairly quick survey at a fraction of the price of in-facility testing.

About this study

A total of 5 shelves were tested. The categories tested provided a good mix of package sizes, number of SKUs on shelf, category purchase cycle frequency, and repertoire vs. impulse categories: pain relief, laundry detergent, gum, wood cleaners, and pasta sauce.

Per category, a total of n=75 interviews were conducted using laser eye-tracking (in-facility), and n=200 using click-tracking technology. Respondents were aged 18+ with quotas to ensure age distribution was proportionate to each category. 70% were female and 100% were New Jersey, USA residents for both in-facility (laser) and online interviews. Other qualifiers included: primary grocery shopper or shared responsibility for shopping, recently purchased items from qualifying category, and standard industry exclusions.

Traditional isn’t always better. Measure what matters.

A pack is the essence of the brand – its role is to introduce, communicate, engage, reinforce, and remind, as well as house the product. These many tasks are given to a relatively small space which needs to activate the target emotional states that drive consumer choice and relationship. The best packaging effectively balances these elements.

Do you know whether it is time to change your packaging? If it is time to change it, are you leveraging both System 1 implicit measurement and aligning it to System 2 explicit measurement to truly understand what to say and convey to drive brand growth?


Voted among the top global insights consultancies for 3 years in a row in the annual Greenbook Research Industry Trends (GRIT) most innovative survey, Hotspex is working with 15 of Top 20 advertisers in over 30 countries because we leverage the most innovative approaches from behavioral sciences, combining System 1 and System 2 measures to truly understand WHY consumer behave the way they do. We then apply marketing sciences, such as the Laws of Growth, to help you find out HOW to apply your insights in an actionable way to build distinct and coherent brands that accelerate growth.

For more information, contact Jonathan La Greca here.

Please share...

One response to “Seeing is not Believing

  1. I’d disagree that your correlations between in-store and attention clicks is good – .67 is pretty mediocre and two of your categories are very poor. If everything was above .80, I’d be impressed. Your correlations between eye-tracking and products purchased are low enough that you could not possibly claim they are substitutions for actual purchase behavior, either in a real store or in a virtual store. Nor could you claim that either version is a good predictor.

    As for eye-tracking being the standard for package design research, that’s hardly the case. It is a useful tool but anyone relying purely on eye-tracking, no matter how it’s done, will make lots of errors. A better package is one that sells more, not one that attracts more attention or gets better attribute ratings.

Join the conversation