A UX challenge: how can we make online shopping with Waitrose easier?
A simple sounding question with many potential answers — this was a big project! But also a great challenge which gave so much room for exploration and research. In this post we’ll explore the steps we took when faced with such a big question.
With such an open-ended question it was important to set boundaries for our work early on. This would ensure we drilled down in the right areas to unearth real tangible benefits for the customer.
Defining the scope
Our work was focussed on improving the online shopping journey from the first interaction that a customer has with the app or waitrose.com through to the order confirmation page, including placing and amending an order. Other customer touchpoints, such as loyalty, would be tackled in a separate workstream.
To help define the remit of our work, we next looked at targets. These were:
- Increase conversion rate
- Increase best customer frequency rate
- Improve customer satisfaction across a number of key areas
We also included principles to help guide the direction of our research and discovery work. This again helped us be specific in our work and not get side tracked as it would be so easy to do considering the range of options available.
The principles:
MAKE IT QUICK
How might we reduce the average time it takes a customer to complete their usual online shop?
MAKE IT SIMPLE
How might we reduce friction through the shopping journey, making the experience as seamless as possible?
MAKE IT CLEAR
How might we reduce complexity — removing distraction, and reducing the number of pages and clicks?
MAKE IT INTUITIVE
How might we reduce mental load — making the online experience easy to navigate for an enjoyable shop?
MAKE IT CONVENIENT
How might we make our customers’ lives easier by automating mundane or repetitive shopping tasks?
MAKE IT PERSONAL
How might we make the experience feel personalised, relevant and tailored to individual customer needs.
The process
We planned out what was required to ensure all aspects of the customer journey were covered, all ideas were explored and all relevant stakeholders were invited to have their say:
- DISCOVERY: We gathered user research, data, and market insights to gain a clear and empathetic understanding of the problems we were trying to solve
- CUSTOMER OBSERVATION: We ran shopping observation sessions on the Waitrose website (desktop & mobile) and app to gain deeper insights into customer perceptions
- CUSTOMER JOURNEY MAP: We synthesised all research and data insights into a map of the core web & app shopping journey, and highlighted the key pain points
- IDEATION AND PRIORITISATION: We gathered as a multidisciplinary team (designers, product managers, researchers) to review the customer journey map and pain points, and ideate a range of potential solutions together
- DOCUMENTATION AND IMPLEMENTATION: We documented each idea as a hypothesis statement with supporting details, and have fed all concepts into product teams to consider for backlogs.
Journey analysis
Now to the core of the project — a chance to really lift the lid on user feeling and attitudes to the end-to-end journey. This was a particularly exciting part of the project for us as a UX team — getting as close as possible to the customer mindset is always the best way to know whether you’re on the right track.
Customer goodwill
The first step was to get an understanding of goodwill towards shopping with Waitrose online. Goodwill is the positive sentiment and feelings of alignment a customer has for a brand. We knew that high levels of goodwill result in very loyal customers, which can improve conversion, frequency and retention. And goodwill can be won by reliably providing an easy, enjoyable and rewarding shopping experience.
But of course there’s a flipside: every time a customer experiences a pain point or is frustrated by the experience, some goodwill is lost. If goodwill levels get too low, the customer is at risk of abandoning a shop or leaving for a competitor.
We undertook an end-to-end audit of shopping journeys to identify pain points. Once complete, we could ideate a range of solutions to help resolve each issue.
Technical errors
Inevitably there was feedback relating to technical errors. These were largely out of our remit but it was important to share these findings with the relevant project managers to ensure they were still being worked upon.
Customer verbatim
We divided each area by workstream and analysed all of the customer comments that we had received over the past. As we get customer feedback from our on-site survey every day, we went through a considerable amount including the good and bad. For example we had comments detailing how easy and even ‘fun’ it was to use our new meal deal builder, while others felt the search function could be improved.
A map for success
We combined all user research and data insights into a customer journey map. This enabled us to build up an understanding of customer sentiment and pain points throughout the shopping journeys.
A ‘Health check’ followed which provided a quick top-level summary of the key issues we uncovered for each area. It was a combination of customer insight, data and our design analysis. We also made sure to apply this insight to web, mobile and app. We were delighted to receive positive comments, but as ever the negative ones had to be our focus. The following is a snapshot of areas we needed to work on:
HOMEPAGE (website)
‘Feels quite on-brand, but layout is a bit blocky and uninspiring, lacks structure, signposting, personalisation and links to key missions.’
SEARCH
‘General perception of poor quality search results with missing items. Lacks ability to easily refine scope on search results page.’
OFFERS
‘Difficult to track what offers are complete, and savings don’t show in trolley or checkout which can be confusing.’
CHECKOUT (app)
‘Having checkout as a web view makes the experience feel quite fragile and not very seamless. Checkout also needs a lot of scrolling.’
Now, next, later…
As we hoped, there were a significant number of changes we wished to implement after such a thorough exploration. Although some stakeholders were keen to have workable solutions that we could implement quickly and easily, we also wanted to pinpoint any long-term improvements that might take more time to develop. To help us do this, we grouped recommendations into now, next and later in order of priority and ease of implementation:
Now — Quick wins that could add value relatively easily
Next — Concepts that might take slightly more effort from development teams, or would benefit from design work
Later — More novel concepts that required further research and testing with customers before making a decision on how to proceed with them
Each suggestion was set out in a logical way with the pain point described in detail followed by our hypothetical fix which laid out what we could do. We then listed the potential value to the business of the fix (such as customer conversion/satisfaction etc). And lastly we labelled what stage the solution was at and the source for the initial pain point (customer feedback etc).
Here’s one example:
AMEND ORDER NUDGE
Pain point: It is quite easy to accidentally start a new order unintentionally, instead of amending an existing order, and there is no way to transfer items from trolley between orders.
Hypothesis: We believe that showing a nudge to customers in very specific scenarios, where it’s likely they intend to amend an order, will increase customer satisfaction and shopping metrics, because customers will be much less likely to accidentally start a new order unintentionally — and will save time and effort.
Potential value: CSAT, conversion, frequency, retention, IPB, AOV
Status: In test queue
Source: Customer verbatim, Internal discovery
(illustrative concept above)
Share for success
The final stages of the project involved putting all our work into a deck to present and share with product managers so they could add the relevant improvements into their respective roadmaps — whether it be for the Browse, Trolley or Checkout teams. They can then be prioritised against existing backlogs to see whether we go ahead with each concept.
I hope you found this approach to such a wide ranging question useful. By taking a simple step by step approach, it’s been really satisfying to see what positive changes can be identified. And it’s certainly opened my eyes to the role of research in the planning stage — rather than the validation phase. This was vital in helping us unearth what customers want to see when shopping with us and pointed our designers in the right direction when creating more relevant and useful experiences.
