The Truth About Tropicana, or How to save $100,000,000 Using Observational Research

Share

Dear Mr. Neil Campbell, President, Tropicana North America

RE: The Truth About Tropicana

We all make mistakes, so no one can blame you for saying that the Tropicana repackaging re-call was a concession to noisy bloggers and social media outcry amongst passionate Tropicana fans. It’s a credible story and one where you appear to have bowed to consumer pressure – something that allows you to use the situation to your advantage – you, Tropicana, have consumers so attached to your old-packaging that you had no choice but to revert back to it, to keep them happy!

However, the truth is much simpler. With the new packaging design Tropicana suffered a $33 million sales drop and without considering properly why this might be the case you risk confusing yourself about the real cause for Tropicana’s sales loss.

You see, while you undoubtedly blame the sales loss on the repackaging, you probably blame it on the wrong elements of the repackaging. And that could prove a fatal mistake. Continue reading

Behavioural Economics and Eye Tracking

Share

I’ve recently read Dan Airley’s excellent book ‘Predictable Irrationality’ and it got me to thinking about eye tracking.

Much of the book is about the importance relativity plays in making decisions; we don’t make decisions in isolation and we don’t make them based on a perfect model of the World. We often make decisions based on the information that we have available at the time we make them.

One of the examples Dan goes into is from a subscription advert for The Economist. The advert gives three options:

  • Economist.com subscription US$59.00
  • Print subscription US $125.00
  • Print & web subscription US $125.00

When given this opportunity 84% of readers chose print and web and 16% choose web only.

He then ran the experiment again with only two options:

  • Economist.com US $59.00
  • Print & web $125.00

The results reversed; 68% chose web only and 32% chose print and web

Dan goes on to talk about the importance of the middle option, even though no one chose it, it helped people figure out what they wanted.

This led me to thinking about how we maybe sub-optimising gondola ends; when we only show promotional items we may be denying the opportunity for the consumer to see the added value the promotion offers when compared to full price items.

I wanted to explore if this was true so I ran an eye tracking experiment with 60 consumers whom I asked to choose an Innocent Smoothie. 30 of the consumers were shown a gondola aisle shelf with only promotional items from the planogram in Figure 1.1.

Fig 1.1 Offer only smoothies

Fig 1.1 Gondola Isle planogram showing only Innocent Smoothies that were on offer.

And 30 consumers saw a section of the chilled isle that contained smoothies on offer next to a non offer smoothie shown in fig 1.2.

Fig 1.2 Offer and non offer smoothies

Fig 1.2 Chilled isle planogram showing Innocent Smoothies on offer and at full price.

The heat map shown below demonstrates the consumers engaged with the promotional offers.

Fig 1.3 Eye Tracked offer only smoothies

Fig 1.3 Eye tracked gondola Isle planogram showing only Innocent Smoothies that were on offer.

Fig 1.4 Eye Tracked offer and non offer smoothies

Fig 1.4 Eye tracked chilled isle planogram showing Innocent Smoothies on offer and at full price. The yellow circle highlights the attention on the full priced smoothie price.

The heatmap shows the consumers engaged with the promotional offer AND with the full price item, not on promotion.

After selecting the smoothie of their choice I asked the consumers if their purchase was good value for money on a 7 point likert scale, 1 being very good value for money and 7 being not very good value for money.

Consumers who saw the promotional item only items gave a mean score of 2.4. Consumers who saw the promotional items next to a full priced premium offer gave 1.7 even though they purchased the same item!

This is similar to another example that Dan mentions in his book; well run restaurants in New York often have a very expensive dish on the menu, its placement is not for it to be sold, but to make the other items on the menu appear to be better value for money.

One of the things that interests me about behavioural economics and eyetracking is that both focus on the decisions people make and offer insight in to how subconscious processes influence their decisions, I hope you do too.

You can see a 20 minute talk on this subject by Dan Airley speaking at the TED network it’s 20 minutes well spent!

As part of Think Eye Tracking’s carbon positive policy each approved comment on this and any other blog article on the site we will pay for a tree to be planted in Carbon Clear’s Sustainable Community Tree Planting India.

Thanks for reading! http://twitter.com/modestrobert

Why task is important in eye tracking research

Share

This is something that is very familiar at academic eye tracking conferences in fact every other paper refers to it, but I don’t ever recall seeing it in the commercial sector.

In 1967 Yarbus eye tracked people viewing the painting “They did not expect him” by Repin. The instructions given to the participants varied from, among other, free examination, to the people’s ages, and how long the visitor had been away.

In the gaze plots below, you will see significant differences depending on the instruction given.

At this point, probably needless to say, when using eye tracking in market research, it is very important to give appropriate instructions to the participants, and when analysing heatmaps to know what instruction was given.

In our ThinkPack eyetracking  test for the FMCG and CPM market, we show participants two planograms, but give them different tasks with each one as we are investigating two different shopping experiences and elements of the pack performance. With the first planogram we are interested in the customer’s free shopping behaviour to see how well the product stands out on the shelf: Do people notice the product? With the second planogram we are interested in how easy it is to find the specific product: Do people confuse the product with another one? Does the product branding stand out?

In the first heatmap below, participants were asked to shop for deodorant. They were to take a look at the shelf and click on the product they would choose to purchase.

In the second heatmap, participant were again asked to shop for deodorant, but to buy “Formula 1 Performance” deodorant as their friend had recommended it. You can see two very different heatmaps are produced.

Just to reiterate the message here; when designing eye tracking research be careful what task you give to the participants, and when drawing conclusions from eye tracking outputs be aware of the task given. It is not easy to control your eyes, but it is actually fairly easy to control what people look at during an eye tracking exercise simply by giving particular tasks. That said, you need to be careful that the outputs are not a result of inappropriate tasks.