Self-checkout: What shoppers want to do is rarely what they end up doing

One of the first things retail executives learn is that shopper surveys are horrible indicators of what shoppers will do in stores

self checkout
Alberto (Creative Commons BY or BY-SA)

One of the first things retail executives learn is that shopper surveys are horrible indicators of what shoppers will do in stores. Asked if they would make purchases at a breached retailer, they’ll routinely say no. But quarterly earnings betray the truth that being breached has just about zero influence on revenue. (Blame zero liability, but that’s a column for another day.)

The issue for today is self-checkout. Surveys show that shoppers love the idea. Retailer experience shows that shoppers don’t love self-checkout the reality nearly as much as they love self-checkout the concept. Reality messes things up, with fruits and vegetables that need weighing and age-restricted products and long lines and filled-to-the-brim shopping carts that were never supposed to be handled at self-checkout. Shoppers are quick to dismiss the value of an associate at a staffed checkout lane, ignoring the fact that their experience of scanning millions of SKUs makes them awfully good at it and impressively fast.

What brings this up is a survey that crossed my desk from a vendor called Interactions. It surveyed 1,011 shoppers and found interesting, contrasting answers. Some 62% of those shoppers said that they prefer to be greeted by “a real person” (as opposed to a fake one?) in-store, but only 10% of them said that they wanted a store associate to play any role in “part of their checkout process.”

Taken literally, that means that 90% of them want associate-less checkout, which is self-checkout. And yet, if you go into any U.S. store that offers a choice of staffed checkout and self-checkout, the percentage of shoppers routinely opting for self-checkout isn’t 90%. Heck, it’s not even close.

Rekha Ramesh, global head of IT and digital strategy at Interactions, agrees that the stated shopper preferences don’t match real shopper behavior, but she considers that normal. “It’s not that they don’t want self-checkout. They clearly do,” Ramesh said. “It’s that it’s not very smooth. It’s not being made easy.” In short, shoppers want self-checkout in a vacuum, but the day-to-day experience doesn’t live up to their hopes. And each bad self-checkout experience leads to more hesitation to try it again.

The reason the experience can be sub-par is that not every basket lends itself equally well to self-checkout systems, most of which were designed to deal with basket sizes of ten or fewer items. If a retailer really wanted to push self-checkout and realize more of those operational savings (from having fewer staffed lanes), it could assign an associate to stand in front of the checkout lanes and encourage/discourage shoppers. Overflowing cart shoppers would be gently guided to a staffed lane, along with customers with age-restricted items or too many weight-charged items, while others could be told, “Your shopping today would work wonderfully in self-checkout.”

In other words, use your associates to train your shoppers when self-checkout is a good idea and when it’s not. That’s a non-technological way to improve the self-checkout experience.

That all said, there is something inherently inefficient in centralized self-checkout. That inefficiency is, in theory, fixed with mobile checkout. Instead of throwing 58 SKUs into the cart and then, at checkout, scanning each item, one at a time, isn’t it far better to scan the barcode in-aisle, right before it’s thrown into the cart? That way, when the last item is thrown in, the checkout is 99% done. Walmart did a trial using that approach several years ago, but the chain has never mentioned it since.

But mobile checkout — which has been the hot retail IT topic for more than seven years — suffers from a serious case of chicken-or-the-egg. A handful of retailers experiment with it, but don’t see enough shopper enthusiasm to merit full deployment. That’s mostly because those shoppers have so little chance to try mobile checkout that they can’t get super comfortable with it. It can’t become a comfortable habit.

Hence: chicken-or-egg. A large number of retailers won’t support it without huge shopper enthusiasm during trials, and shoppers won’t deliver that acceptance until there are far more retailers offering it. Need a more practical example? Look at Apple Pay’s plateaued share of U.S. transactions.

The most daring experiment is from Amazon's Go stores, which bypass barcode scanning entirely, replacing it with a series of cameras and video analytics. This store has many practical problems to overcome — Forrester analyst Brendan Witcher said it won’t take long before college students figure out that they can “create a huddle and the store will have no idea what they just grabbed” — but the biggest issue is that it’s stupendously expensive.

It works well as a concept store, but the idea that Amazon could successfully license and sell this to other retailers doesn’t seem realistic. “I think what Amazon is doing is the right idea,” Witcher said. “The question is viability.”

But Witcher’s thoughts on self-checkout reinforced Remesh’s thoughts. “Consumers often believe that it will be convenient” even though it often isn’t, Witcher said.

Shoppers today think they’ll really like self-checkout. Now, if they only did.

Download the 2018 Best Places to Work in IT special report
  
Shop Tech Products at Amazon