Improving the In-Store Colleague SmartShop Experience

UI/UX Update
June 2019 - March 2020

With the introduction of SmartShop on both the handset device and the mobile app, there was a concern in how loss prevention (stealing) could be managed with a product that pushes autonomy as a selling point.

So as part of the SmartShop experience, colleagues were trained in how to perform interventions, otherwise known as rescans. SmartShop on handsets, unlike the app, has been available since 2016 - so the rescan process has been around since then.

With that said, the UX/UI hadn’t been touched for the colleague experience since then either. As this was my very first project when I joined Sainsbury’s in June 2019, I had no idea what rescans were, how they worked or when they were administered. The main brief for the update was to reskin the old UI to be fitting with Sainsbury’s Luna guidelines (the Sainsbury’s UI library) as a first step, and then to also consider how age verification would be performed. At this time in June 2019, customers that used SmartShop could not purchase alcohol - so this was the other side of the coin to think about.

First things first, what’s a rescan?

A rescan is a process where a SmartShop customer is randomly selected based on the fact that it's their 3rd or 5th shop. The rescan is conducted by a Sainsbury's colleague, and all rescans begin as a 'partial' rescan. This means that only 10% of their trolley or basket will be scanned - if anything is detected on the colleague handset device that is not in the customer handset/mobile app - then a full rescan commences to check the entire customer trolley. This could be over 100 items.

All rescans are conducted within the SmartShop area, which is small area busy with other customers, trolleys and self checkouts.


What did the old colleague device UI look like?

The previous UI for the colleague app/handset was not in line with the Sainsbury's UI guidelines, so my main task was to look at this all over again and update it from things like the input fields to various font sizes. Of course, while I was looking at this I was also looking at the overall flow and thinking of how we would eventually slot in another feature, age verification. At this time, age verification did not exist for colleagues on their app, so verifying customer ages had to be conducted through self-checkouts.


How was it improved?

The main improvements were offering stronger indications to colleagues what to do next - and a visible countdown of how many more items they needed to scan in order to fulfill the partial rescan quota of 10% of the customer shop.


How will colleagues verify customer age via this app?

To tag onto the rescans function in the colleague app, colleagues would also need to conduct age verification from their handset. Since this didn’t exist at all, I wanted to understand what the existing age verification proceses were - and if there were anything that we had to ask for legal reasons (i.e., reasons for rejecting a sale, etc).


Across the traditional tills, self checkouts and tobbaco & lottery tills - the same type of reasons were given for rejecting customers (i.e., they’re inebriated, they’re underage, they don’t have an ID with them).

My next concern was how to introduce age verification into the colleague app, and to consider scenarios like: What if a customer triggers both a rescan and an age verification? How can I make this as painless as possible for both customers and colleagues?

After some back and forth on flows, prototyping a few versions and testing to see which felt the smoothest - I came up with the following flows - this fits for both customers using the SmartShop and handset:


Testing in-store


After discussing the flow and prototyping what it would be like for a colleague to perform a rescan and age verification (and testing in the offic while role playing with my office colleagues) I prepared to test in-store. Myself and a product owner visited Nine Elms Sainsbury's and tested the updated colleague 'app' on handset with four colleagues.

It was a little tense, because these store colleagues were still on the clock so I wanted to ensure that this process was as quick and easy as possible. Thankfully, during the testing the feedback was positive on the updated rescan process and new age verification process. However - they did bring up questions like how are we going to make this even 'smarter' - will customers eventually just be able to take a selfie or use their drivers licence with the scan technology on their phone so that they no longer have to perform interventions - which were all great questions.

What’s happened since this UI/UX update?


Customers can now purchase alcohol, and rescans and age verifications can be performed across customers that are using handsets and the app. Having said that - ideally, neither one of these processes would be necessary.

Having a proposition that promotes autonomy and speed is at odds with processes that are intentionally interrupting a customer flow - and it’s been a massive pain point for the design team to contend with. It’s important to protect profit, but it causes customers to feel that they’re doing something wrong.

Instead of it being something that’s being miscontrued in the app or handset, it’s down to the environment and being ‘selected’ in front of other customers and going through an abnormal process than if they had checked out via the traditional tills. Right now, only 60% of SmartShop customers are returning to use SmartShop after experiencing a rescan.

A little over a year on from when I designed this process and tested it with our store colleagues, we’re finally getting the chance to re-evaluate the entire service design of rescans. A designer in my team is beginning to unravel the psychology between customers, rescans, the format of the SmartShop area and how colleagues have been coping with customer frustration.

In addition, we’re also looking at the algorithm to be more intelligent to understand exactly why customers should be targeted. Rather than just because it’s their 3rd or 5th time - could it be because when customers purchase multiples of one product, they’re more likely to forget to scan one? So...we keep pushing on to make this better in every way! .