Redesigning SmartShop Shopping Lists
Thousands of users benefited from the new SmartShop shopping list improvements.
SmartShop has a simple shopping list feature where people can add items to their list and tick them off. On the app people aren’t able to share lists or create multiple lists. Around 100,000 people are using the SmartShop shopping list feature each month.
The team wanted to improve the shopping list experience for people. I started by reviewing all previous surveys and usability testing that had been done on SmartShop. I identified 3 types of list behaviour. These list behaviour types are more useful than personas, which can stereotype our customers. I carried out further quantitative research with people to check whether these list types were accurate.
“A few cupboard things but food I like to go in and see what’s there at the time.” User 7, Usability testing Nov — Dec 2019
This person shops in Sainsbury’s local loads of times a week, they shop for a couple of days and grab what they fancy. They are still interested in offers. They may live alone: ”I don’t tend to make lists because it’s only myself I’m cooking for.” Participant 3. 26 out of 100 survey participants said they don’t make a list.
“We’ll write a shopping list …we’re in the store and something catches your eye — we tried the fajitas the other night.” Accessibility User 1, Usability testing Nov — Dec 2019
They do a big shop and get household essentials, and they also look out for extra things they fancy. They have a list, but they aren’t afraid to deviate from it. Their list is either in Notes on their phone or paper. They may have a partner that they share a list with.
“I go by the list. I buy what I’m told.” Participant 6, usability testing 2020
This person is a pro lister. They create a meal plan for them, and others that they live with. They organise their list into categories. They may email themselves items when they remember. They look into the cupboards before they go to see what needs topping up. They encourage family members to contribute to the list on the fridge.
I used these list behaviours to develop 9 ideas for new features. I carried out a card sort with 100 people to see which would be most popular. Participants sorted the features into: ‘would find this useful, wouldn’t find this useful, might find this useful’. They also gave comments on which features stood out to them as being particularly helpful.
The comments were useful — people had a lot to say about auto-tick lists and categorising their shopping list.
Auto-tick lists did very well with only 2 out of 100 participants saying they wouldn’t use it. 84 people felt it would be useful.
I used the information from the card sort to further develop key features. Some features could be tested via UserZoom, the software we use for remote usability testing. The more complex features needed to be tested with live usability testing sessions with SmartShop users.
I developed a Figma prototype with; the suggested items, categories, and auto-tick list features. The prototype also included an updated user interface, with the ability to name a list, and create multiple lists.
I carried out usability testing session over video calls with 8 participants. I asked people about their shopping habits, to understand how likely they were to use the SmartShop shopping list features.
“I’ve got a folder that I built up over the years of recipes that I like.” Participant 7.
The biggest learning from this research is that people’s shopping list habits are very ingrained. 7 out of the 8 participants we spoke to shopped with a shopping list. I ran a survey with a larger sample size: 74 out of 100 participants create a list when food shopping. Out of those who do shop with a list, 6 of our usability testing participants categorised their list according to shopping list aisles. “It’s not as regimented as aisles but I could do that. I’ve got fruit dairy household stuff.” Participant 5 on categorisation. This behaviour was backed up by the quantitative survey data: 30 out of 74 people categorised their list.
Participants were likely to create a paper list and put it on the fridge, encouraging family members to add to the list. Other people created lists within their Notes app, or used a spreadsheet to create a list.
People had their own personal habits and methods around shopping list creation, which means if we develop new features it may not sway users to use them: “I would reluctantly create a digital list and I would also… I would create my list.” Participant 5.
Participant 1 said: “I like the fact that when you’ve scanned them it takes it off your list.” Participants understood how the auto-tick list worked.
However, participants didn’t realise that you could swipe up to view items from the scanner view. This feature would be expensive to implement, and participants were already set in the current list behaviour.
“Helpful feature to remind you what you’ve missed and add them on easily.” Participant 3. Overall people had no trouble with the task of adding coriander to the shopping list, and they liked the notification that it came with. “It seems user friendly and easy to understand I’m getting detailed descriptions of the suggestions” Participant 1.
The trickiest reason that a person might be put off shopping with SmartShop is the fear that someone will think they are stealing. “I felt like I was going getting watched by the security guard on it. It made me feel like I don’t want to do it again.” Participant 2. Due to this we are looking into redesigning our rescan experience, when a customer has their basket double checked by a member of staff.
I’ve only included the key insights here, because I could talk all day about people’s fascinating shopping list behaviour. Some people were elaborate: creating spreadsheets and shopping at multiple stores to get the best deals and products. Some people kept it simple just buying food for their next couple of meals with no list.
We used these insights to further develop and implement the suggested items feature, and make some changes to the user interface.
This feature resonated with people, everyone liked the idea of being able to quickly add in basics and things they’d bought before. We changed it so that people know where the data comes from, a sentence explaining that these suggestions are based on previous items bought. Our engineers are working on an algorithm which will give people meaningful suggestions based on frequently bought items. We have learned from the successful ‘Favourites’ feature on Sainsbury’s groceries online website.
One thing we tested was having 6 dots on the right side of the list item to indicate that the list item is moveable. This is a common UI pattern. However, it tested badly, not a single participant recognised that the list item was moveable. We are removing the 6 dots. In the original shopping list feature the tick is orange, and the font is the Sainsbury’s Maryanne. We are simplifying the look with a grey tick and Arial font — both of these changes are better for accessibility.
This is the new and updated information architecture of the app.
The updated user interface and the suggested items will be rolled out to 100,000+ SmartShop mobile app users in 2020 or early 2021. The team is also updating the menu so that the list function is easier and clearer to access, being more prominent on the home page. We will continue to monitor analytics and user feedback on the shopping list function.
