Design Thinking | Rethinking the Experience of Supermarket Self-Checkouts
Going Digital is a rapidly growing need for the world, especially so when the COVID-19 situation is affecting our daily lives today. Singapore’s digitalisation efforts in supermarkets have introduced self-checkout counters as an alternate option to the traditional counter cashiers. This initiative helps to tackle the labour crunch and cut waiting time for customers, however, is the self-checkout experience really an improvement to the customer’s shopping experience?
In this case study, our team aims to use Design Thinking to rethink the product life cycle and user experience of the self-checkout machines in a local supermarket chain. By applying relevant User Experience (UX) methodologies, we aim to identify key design opportunities and enhance the user experience for grocery shopping in these supermarkets.
This project was done with a team of 5.
01 | Introduction
Most consumers find the self-checkout system highly beneficial. With an automated system, companies can save on various costs and wait times are reduced. Shoppers also gain certain value from taking control of the transaction by being able to scan their own goods and pack them the way they want. This sense of control can lead to greater consumer satisfaction and their wish to re-use self-service technology.
“We have constantly looked for innovative ways to alleviate industry-wide retail challenges such as a shortage of manpower as well as ways to enhance customers’ in-store shopping experience.” Mr Seah Kian Peng, CEO (Singapore), NTUC FairPrice
This UX study will mainly focus on rethinking the product life cycle and user experience of the self-checkout machines in local supermarket chains (we have chosen NTUC FairPrice). We based the research process on Design Thinking in our case study, as well as relevant UX methodologies to empathise with different groups of users, define the problems that they currently face, and come up with key design opportunities to enhance the overall user experience for self-checkout over the spans of 4 weeks.
02 | Design Thinking
Design thinking helps us understand the users, challenge assumptions, and redefine problems, in an attempt to identify alternative strategies and solutions that might not be instantly apparent with our initial level of understanding. We break down the process in the following:
Empathize
We drew insights through shadowing, fly-on-the-wall and contextual inquiry. With that, we created empathy maps for all our users. These empathy maps captured what the users did, said, think and felt.
Define
With our existing empathy maps, we reflected on our collective knowledge through an aggregated empathy map. We analysed the insights we discovered, mapped the pain points and identified challenges faced by the users.
Ideate
Challenged ourselves through the “How-Might-We” (HMW) method to improve the existing system and resolve the challenges faced by our users. We take the ideas further through different brainstorming methods like lotus blossom, scamper and worst idea possible.
Prototype
Created a paper prototype (low-fidelity), after numerous revisions internally, we created a digital version of our paper prototype in Marvel so users are able to navigate and also feedback on our ideas.
Test
We gathered 6 users, 3 from each personas to test on our prototype. We gather their reviews and insights through shadowing & contextual inquiry. Together with all the insights, we organised and evaluated with the “Rose, Buds and Thorns” method.
03 | Empathize
“To find ideas, find problems. To find problems, talk to people.”
Julie Zhou, former VP of Product Design at Facebook, author of The Making of a Manager
Deep and thorough user research is critical to properly define design problems. To see the world through our user’s eyes, our team deployed ourselves at 4 Fairprice outlets to research the experiences of 10 different users, ourselves included, using the self-checkout counters. As a team, we saw the benefit in drawing insight from multiple angles, hence we employed “Immersion”, “Fly-On-The-Wall”, “Shadowing” and “Contextual Inquiry” observations methods on our subjects, before we synergised our observations and insights into an aggregated empathy map from which we draw insights from what they said, did, feel and thought.
Initial Aggregated Empathy Map
Withholding judgement, we collated our raw observations; colour coded to each user, and separated them into what they did, said, felt and thought.
We grouped common comments mentioned into categories which helped us identify the pain points and expectations of our users.
Final Aggregated Empathy Map
Thereafter, we focused on the common points that many users seem to face, grouped common themes into concise and meaningful clusters that provide insights into user behaviour, attitude, preferences, needs and pain points.
Other than the key insights mentioned above, we were encouraged to know that there was sufficient positive feedback to indicate that users who were familiar with the self-checkout kiosks found it convenient and easy to use.
04 | Define
Personas add the human touch to our typically cold insights and research. From our observations and insights gathered, we created two unique personas that best represented the users in our research. Through the personas, we were able to better understand our users’ needs, experiences, behaviours and goals.
Persona 1 | Jeremy
Jeremy represents tech-savvy users who are familiar with self-checkout counters. He has high expectations when it comes to speed and efficiency.
We visualised Jeremy’s shopping behaviour and touch-points in the supermarket, and empathised with his pain points. With that knowledge, we were able to identify possible design opportunities to improve his checkout experience, and summarised it in a Customer Journey Map.
Jeremy’s Customer Journey Map
Through his journey map, we identified that tech-savvy users are comfortable with the use of self-checkout counters, and that they prefer to do so.
Jeremy’s pain points:
1. Self-checkout queue: Long queues
2. Card/Scanning Errors: Might require external help from staff to help rectify the mistake. It would require extra time spent during the process.
3. Bagging Area: Plastic bags are not readily available in some outlets visited. Some users do not bring their own bags and require that the plastic bags are accessible at their station.
4. Receipt: Upon payment, users can’t get their printed receipt, not knowing that they actually skipped the print receipt option in the process.
5. Can’t find the things that they need in the process
Persona 2 | Patricia
Patricia represents users who are newer to technology, but are optimistic about learning and applying new knowledge to make life easier. They unfortunately belong to a user group that falls through the cracks of the digital transformation, and have already formed deep habits of how things should usually work. Patricia is eager, but she needs help when learning this new self-checkout process.
Patricia’s Journey Map:
Patricia is vastly different from Jeremy. She struggles with the self-checkout process but, motivated by the promise of efficiency, she tries to embrace the changes.
Patricia’s pain points:
1. Pressure: Users are very aware of the queue behind them, making them stressed and afraid of holding up the queue. In addition, all the errors compound their frustration, increasing the learning curve.
2. Confusion: Not knowing what to do, because of confusing instructions.
05 | Ideate
From the pain points identified through our personas, we focus to address the following concerns:
1. Simplify the interface for new and current users
2. Ease the pressure received from the queue
With the “How-Might-We” (HMW) method, we brainstormed initial ideas of how we may address the pain points.
We took the ideas further through different brainstorming methods like “SCAMPER” and “Worst Idea Possible”.
With all the ideas explored, we arrived at the decision to break down the process for our users in 3 stages.
The 3 stages can be a challenge at any stage for some users and could easily be the reason behind the long queue and growing pressures for users. By breaking down the stages, users can deal with it at their own pace.
06 | Prototype
1. Scan
We discovered a smart basket used in Japan. With the use of RFID tag, the items are automatically scanned the minute users dropped into their basket.
“We need a good solution also for customers who wouldn’t like a completely digitalized system,”
Yasuyuki Fukui, a Panasonic business-development executive
This smart basket benefits users of all age groups as you do not need to be tech savvy to know how to operate it. It works just like your regular basket but helps resolve the scanning issues and time spent on scanning your items.
2. Pay
We developed a simpler interface for our users at the self checkout kiosk to resolve the following issues:
• for members to identified themselves to the system
• for members to redeem their linkpoints
• for use of vouchers within the kiosk
• to include only straightforward concise informations that users require
With the current Covid-19 situation, we want to be able to discuss as a team without meeting up and have a rough idea of the functions to include in our prototype, we started working together from google slides in the beginning just to get everybody on the same page:
Afterwhich, we started with a paper prototype to experiment with our ideas:
After some edits, we moved on to upload our ideas from paper prototype onto Marvel so that we can conveniently test it out among the team and also allow users to try navigating with it.
3. Pack
Self packing is becoming common, brands like Donki, Uniqlo, Daiso are all implementing stations for users to pack at their own pace so the next user can immediately make payment once you are done with yours.
07 | Test
We did the test with a total of 6 users consisting of 3 Jeremys and 3 Patricias based on our personas. We explain to the users that scanning was already done through the smart basket and their task is to make payment through the prototype we created in Marvel.
During the test, users are tasked to check out as members, required to use their linkpoint and vouchers. For some users, they may experience a scenario when they want to add a new item in the process. We break down our insights with the “Rose, bud & thorns” method.
We organised the insights into the following:
08 | Conclusion
The response we received from our users was positive, answering their pain points which is a good start for our initial assessment. Some issues that we need to address for further development:
1. Educating users
Introducing and educating users to help them gain confidence with the smart basket. This could be done in the beginning the minute they picked up the basket in store.
2. Users confused by the voucher
Our feedback for the voucher buttons was met with mixed reviews, however we also had to take note that the current self-checkout machines do not cater to voucher use, and therefore this is something new to the user as well. This will therefore take time for users to learn and get used to the new feature, so it is expected for users to be slightly confused.
This could be due to the position of the button, the color of the button, the timing it appeared. As suggested by the user, maybe the voucher option should appear when they are about to make payment. The appearance of the button can be explored in future development. We will have to look into enhancing the user experience and experiment on what users expect to see, hear, and feel for this function.
3. User interface related problem
With our paper prototype being digitalised, users have higher expectations of how it looks as compared to how it works. Issues that were brought up were issues that can be addressed in the future development when we are building a high-fidelity prototype.
4. Loophole
User pointed out a loophole of how fruits and vegetables get taken without scanning as these are items that require weighing before scanning.
We focus on addressing the pain points and proposing the possible solutions to it but as implementation of technology is not immediate, this is our long shot yet daring attempt on the issues.
Final Reflections
Overall, while there were more improvements to be made specific to the UI visuals of the self checkout screens, we did manage to address our initial issue of reducing the stress of customers who want to use the self checkout machine, catering to the pain points of two completely different target users.
Instructions could also be further improved on, whereby we continued to make small iterations to our paper prototype digitally so that we may address these issues.
However, we also had to realise that many other factors would affect the user experience outside of just visuals, for example sound, if there were staffs still visible around the area, if users would take a while to adjust to the new system, and so on.
While we come up with new user experiences to potentially enhance and better improve the quality of life, we must also realise that there is always a learning curve to overcome and it differs between people. At the end of the day, as experience designers we also have to keep in mind that users who have a reluctance to learn are also users we have to cater for, and thus an idea with huge novelty will also have to go in steps to ease out the learning curve.
