You Can’t Sell What Shoppers Can’t See: A Register-Level Insight Story

In honor of National Data Science Day, I thought I would do a shout out to all the insights professionals I have been lucky enough to work with over the years by asking a simple question. What is the coolest insight you have ever uncovered?

Here’s my favorite insight from a few years back—about the time grocery self-checkout registers were becoming more popular.

Have you ever noticed that some cash registers at your local grocery store never seem to be open? Some lanes are semi-permanently blocked by corrugate displays or rolling metal racks. That odd sight sparked a hypothesis when I worked at the Wrigley Company: what if we could improve sales and reduce waste by moving gum, candy, and mints from the closed registers to the ones that were actually in use?

It made sense. Moving product from a dormant lane to a high-traffic one would take effort, but it could reduce out-of-code waste and increase in-stock rates at the open registers. So we asked Safeway for register-level point-of-sale (POS) data to test our theory.

That’s when everything changed.

Once we had access to the data, we discovered something so unexpected that we thought it had to be a mistake.

The register level data—stripped of shopper PII—contained sales of gum, candy, and mints by register, along with the total transaction value. We began parsing millions of transactions from 10–20 registers across 1,200 stores. At the time, Safeway told us no CPG vendor had ever asked for register-level POS data.

What we expected to see were patterns of inactivity from the “always closed” lanes. And yes, some registers showed zero transactions for an entire month. But then we noticed something else—some registers were very busy. A few showed transactions continuously, 24 hours a day, 7 days a week.

Again, this may not seem odd today in the era of self-checkouts. But this was prior to self-checkout becoming ubiquitous—and the scale of it was astonishing. A quick regression showed that in 800 of the 1,200 stores, a single register was responsible for nearly half of all transactions. Just one.

If you don’t believe it, next time you’re in a grocery store, count how many registers are staffed. Outside of peak hours, it’s often just one or two.

So why did this matter?

We went deeper. We analyzed every transaction—whether or not it included a candy sale. And what we found floored us. In half of those 800 stores, the busiest register had virtually no sales of single-serve gum, mints, or candy bars.

We assumed the data had to be wrong. We asked our partners to rerun the extract. Then we asked them to rerun it again. Same result.

Despite their traffic, the busiest registers weren’t selling their fair share of impulse items.

So we hit the road.

We visited 30 stores across five states to observe shoppers in action. And there it was: in nearly every case, the high-volume register was all the way at the end of the row—and it was often the “10 items or less” express lane. These registers had no gum, candy, or mint merchandising racks.

Because these items are impulse purchases, if shoppers don’t see them, they don’t buy them. And if the register isn’t merchandised, shoppers don’t typically walk to another one that is.

We’d stumbled upon a massive blind spot in merchandising execution.

Now we had a real insight. But insights only matter if you act on them.

So we built a quick business case. If we could ensure that every shopper checked out at a lane with candy, gum, and mints, we estimated an incremental $3.5 million in annual sales. That made this insight not just interesting—it made it valuable.

As the Wrigley Sales Leader, I took it personally that we were missing this opportunity. Together with category and store operations teams, we moved fast and implemented three corrective actions:

1. Field sales teams worked in-store to install over-the-belt racks for gum, mints, and candy at all staffed and self-checkout lanes.
2. In stores where over-the-belt space was limited, we added combination racks to the left/right of express lanes.
3. For stores where racking wasn’t feasible, we sent individualized emails to store managers, highlighting their store’s data and the upside of choosing to open merchandised registers first.

We moved from insight to action in under three months—and the impact was immediate. We drove over $1 million in incremental sales in the first year and reached nearly $3.5 million annually in subsequent years.

The proof? Fifteen years later, when I walk into those same stores, I still see fully merchandized candy racks at the busiest lanes that we proposed.

That’s when you know your insight had staying power. You didn’t just find a pattern—you changed behavior.

This experience taught me a few lasting lessons:

• If something looks strange in the data, dig deeper. Anomalies are often hidden opportunities. • Quantitative and observational insights are stronger together. Sometimes you need to get out from behind the spreadsheet. • Insights don’t drive value until they’re activated. Quantify the upside. Build the case. Take action. • Always stay curious. We started with a sustainability idea—and wound up reshaping an entire merchandising approach.

Now it’s your turn. If you work in: • Shopper or Consumer Insights • Retail Analytics • In-store Execution • Trade Marketing • Data Science for CPG

…I want to hear your story.

What’s the most surprising, game-changing, or just plain funny insight you’ve uncovered? Did it shift strategy? Change the shopper experience? Deliver measurable impact?

Drop it in the comments. Or post your own story and tag me. Let’s bring the best of our industry’s thinking into the spotlight.

Because in the end, insights only matter when we do something with them.

#ShopperInsights #ConsumerInsights #RetailAnalytics #InStoreExecution #CategoryManagement #CPG #Retail #ThinkingBeyondTheStoreVisit #InsightActivation #ImpulseSales #DataDriven

Originally published on Jul 22, 2025