Steering without engineering
Merchandisers needed to see and adjust why products ranked where they did, without raising an engineering ticket.
Carhartt WIP, 2026
I own Carhartt WIP's move to a new discovery platform and its scale-up: the requirements, the vendor choice, the roadmap, the UX, and the scrum team of engineers delivering it on web and app.
The problem
Merchandisers needed to see and adjust why products ranked where they did, without raising an engineering ticket.
The team wanted to set up and read tests on its own.
The engine needed product data straight from the source catalog rather than through a separate feed.
What I did
I wrote 87 requirements across search, merchandising, recommendations and the platform itself, and split them into MVP and later phases. Every demo and every vendor score used them, which kept the decision about our problems rather than the best demo.
Two-week sprints from contract to MVP. I own the board, the priorities, the measurement strategy and the merchandising rules, and I set the go-live plan across web, app, the implementation partner and the vendor.
Attribute names now match the source catalog, and new-arrival dates now stay stable, so the new engine starts clean.
Curated filters per category, six to nine per page. Merchandising rules sit on top of the ranking model, so campaigns survive model updates, and sold-out products stay visible but rank low.
I audited every existing placement and set one principle for all of them: each placement gets one goal, a primary strategy, a fallback and guardrails. Search keeps sold-out items ranked low, while recommendations leave them out.
The MVP is live in Germany. I own the plan after it: rolling out market by market, adding recommendations, and testing the team can run on its own.
Where it stands
I'd set the pre-launch baseline even earlier, during requirements, so every phase is measured from the same starting point.