2024-25 / B2C/ Q-Commerce / iOS - Android - Web
Improving the experience for customers after they've placed an order and laying the groundwork for order editing and out-of-hours ordering
18% reduction
in repeat inquiries
16% fewer
support contacts
We had some analytics tracking the paths users took on their journey, customer support data in Intercom, and shared anecdotal knowledge amongst the team and stakeholders. This painted a picture of a number of issues, but two flows had the greatest negative impact on customer experience and subsequently customer support workflows.
Digging Deeper
I audited the existing help centre — mapping every entry point, flow, and drop-off across the app — and cross-referenced it with support ticket data from Customer Operations to connect failure points to specific UX problems.
From there, I designed the user testing plan we could implement to get more first-hand input from customers.
I used Google NotebookLM collate all of the data from our user research, customer support tickets, and competitor desk research into a single project we could query for insights, summarise data, and help produce product artefacts for the team and stakeholders.
Alongside NotebookLM I used ChatGPT to help with UX copy suggestions, quick audits of design solutions.
Refined insights
Operations were working to a deflection metric which incentivised keeping customers from interacting with agents. Result = contacts bubble up through other channels and poor reviews.
Current reporting issues flow was useful but lacked guardrails and clarity.
Not every customer query was unique and many were a chain of initial question followed by status checks.
Hypotheses
If we place help content upfront and try to answer questions early then support contacts and customer frustation can be minimised.
If we match the language to how people actually frame their issues, they'll find what they need and contact us less.
If we give customers clear confirmation and status after reporting an issue, we'll prevent duplicate submissions and reduce contacts by users checking the status of a claim.
Solutions
Help Centre Hub
Changes to the first help touchpoint were designed around some key aims:
Match verbiage on the page to the content we show
Improve visibility of the all orders that a customer might require help with
Surface useful content that is either currently only in FAQs or can only be accessed via the Intercom flow
Help topics were hidden inside the Intercom flow, structured as a tree. This let customers navigate to their specific issue instead of hitting dead ends and looping back through the bot.
If the answer still wasn't enough, they could tap through to an agent. That handoff carried a tag showing exactly what topic they'd been on, so agents saw the context immediately.
Order number
Customer name
Question Lvl 1
Question Lvl 2
Issues With Items
We redesigned the Issues with Items flow to clarify the experience, reduce errors, and prevent duplicate submissions.
Key changes:
Unique wording for each step
Persistent issue tags
Quantity input for multiple instances
Specific CTA naming
You could previously report the same item more than once — handy if it was both missing and damaged, but most of the time it just created duplicates. We block that now. Cleaner for the majority, occasional edge case trade-off.
Status & Progress Visibility
Many contacts weren't new problems. Customers didn't know where their order or refund stood, or whether they were eligible. Email engagement was low, so refund updates were often missed.
I redesigned how status was communicated. I explored changes across in-app, email and push, and reworked the UX copy for each state. Order status states now flag action, share updates and point customers to email for detail. Support flow refinements cut uncertainty-driven contacts, contributing to an 18% drop in repeat contacts across the Help Centre work. Tying in-app signals to coordinated email became a cross-team initiative beyond the help centre scope.
15 April 2023
Delivered
15 items
Frankfurter Tor 4
+9
items
Item report processed. Check your email
for details.
Item report submitted. We’ll send you an email within 24 hours.
Refund successfully processed! Check your email
for details.
Outcomes
Support contacts dropped 16% after we rebuilt the Help Centre around customer intent. Customers found answers themselves instead of raising a ticket.
Repeat contacts fell 18% once contextual help sat at the point of reporting. Customers weren't left guessing what to do next.
CSAT rose 2.7 points for customers who used the Help Centre.
