Decathlon
E-commerce Technology PM Intern · 2025–2026
Decathlon's China e-commerce team runs across WeChat Mini Program, Douyin, Tmall, and Taobao Flash Sale simultaneously. The PM team sits between business and IT — they gather requirements, translate them into specs, coordinate development, and manage testing before launch.
I was embedded in this team for three months. My mentor gave me an open brief — learn the system, find problems, contribute. That turned out to define how the internship went.
WeChat Mini Program UX
Nobody assigned me this. In the first few weeks, I spent time using the Decathlon WeChat Mini Program as a regular customer would — browsing, adding to cart, checking logistics, going through the returns flow. Things kept bothering me, so I started writing them down. When I had enough observations to make a structured case, I brought it to my mentor. His response: if you think it's worth pursuing, build the analysis and we'll see if it justifies a project proposal.
No internal user behavior data was available — no heatmaps, no conversion funnel breakdowns. I flagged this to my mentor; his judgment was that the pain points were visible enough to proceed without it. So I competitive-benchmarked instead: full-funnel walkthroughs of JD 1P, JD 3P, and Tmall, covering the product detail page, logistics tracking, and the after-sales and returns flow. Documented everything with screenshots and written notes.
Identified 16 pain points in total. Key ones: SKU naming inconsistency between the Mini Program and off-platform content — breaking the seeding-to-conversion flow; missing logistics ETA display — a low-cost fix with high customer confidence impact; no self-service return with courier pickup — customers had to contact customer service manually while competitors had fully automated this.
With 16 issues and limited dev bandwidth, I needed a prioritization framework that both I and the team could agree on. Built a Value-Effort matrix with dual-dimension weighted scoring — consumer need intensity (0–5) and business value (0–5). Ran two alignment meetings with the team's PM to validate quadrant placements. High-priority items went into the sprint; others into the backlog.
My mentor initially wanted to prioritize self-service returns with courier pickup. I pushed back — not on the goal, on timing. The feature requires reverse logistics API integration and courier partnership agreements; short-term build cost is high, and adding a frictionless return path would likely spike return rates before those costs are recouped. We agreed it belonged in the long-term roadmap, not the current sprint.
UAT Coordination
While the UX project was running, I coordinated UAT across about eight concurrent projects — spanning DTC manual refunds, two new Douyin stores (kids + apparel/footwear), Tmall, and Taobao Flash Sale. The role was essentially information routing: track what tests are scheduled, record issues found, get business, IT, finance, and customer service aligned, and follow up until every issue was closed.
On the DTC manual refund project, Round 1 testing was done and IT had completed their fixes. The customer service team assigned to Round 2 re-testing went past deadline with no response to WeChat messages. I decided async was not working. Walked over and scheduled a short in-person meeting — but did not open with "you're overdue." I asked about their situation first, then explained what the project would actually do for them: standardizing the refund process would reduce their compliance exposure and cut the manual workload they dealt with on every price-difference case. They had not thought about it from that angle. Re-testing was completed promptly afterward.
When cross-team alignment stalls, the problem is usually one of incentives, not communication channel. Switching from async to in-person was useful, but more useful was leading with "here's what this does for you" rather than "here's what you're supposed to do by when."
PM Tool Optimization
My mentor, knowing I was studying project management, assigned me to benchmark the team's internal PM tool against what good PM software actually does. Decathlon was building its own rather than buying — the logic being that their specific workflow did not map cleanly onto Jira or Monday.
Benchmarked against Jira, Trello, Monday, DingTalk, and Feishu. Identified 13 pain points in the internal tool. Key ones: no project progress visibility without clicking through to an external Jira link; no clear owner mapping — could not tell who was responsible for a project without asking around; dashboard number cards lacking context — just a number, no trend or comparison; milestone hover states not showing detail; filter logic inconsistencies across select-all, reset, and multi-select.
Wrote up a complete requirements iteration document and aligned it with my manager. Separately noticed the team was using Notion for team-level task tracking and finding it insufficient — free tier limitations, not designed for kanban workflows. Proposed Trello as a lightweight complement — not a replacement for the internal tool, just the right tool for the team's day-to-day. The team adopted it and continued using it.
Reflection
The UX project is the one I'd spend more time on if I did this again. Competitive benchmarking was useful, but I should have pushed harder to get internal data — even partial conversion funnel data would have strengthened the priority arguments significantly. "The pain points are obvious enough" is a reasonable pragmatic call, but it leaves the prioritization open to challenge.
The UAT coordination work was valuable in a different way. Managing eight concurrent projects across four or five stakeholder groups in a large organization is messier than it looks on a Gantt chart. Most of the actual work was keeping information current and making sure the right people knew what they needed to know. When that breaks down, the fix is almost never "send a clearer message." It is usually about changing the frame.