Sr Manager, UX Product Design
North America’s Largest Apparel Platform.
I built the UX, UI, and research practice here, from one designer to seven, and made AI part of how the team works. An agent I built cut launch-catalog writing from weeks to under an hour.
I built and still lead this practice, so the product and process detail stays private. What’s public is the team, the decisions, and the outcomes. Unlock
My designers own the hard problems. I build what makes them faster.
- Outcome
- Most of the roadmap starts as work my team incubates, and one opportunity they surfaced became a top executive priority
- Shipped a design tool now used by thousands of customers and staff across the platform
- Built an agent that reduced launch-catalog writing labor from weeks to under an hour
- Helped stand up AI governance and a team adoption model across the company
- Problem
- North America's largest apparel supplier and distributor, a multi-billion-dollar B2B and B2C platform whose design org was producing slower than the business moved, with research and exploration treated as luxuries instead of defaults.
- My role
- Built and lead a seven-person UX, UI, and research practice, grown from one designer
- Lead UX, UI, and research for a multi-billion-dollar B2B and B2C platform, partnering with product management
- Drive AI adoption across the company, working with our AI enablement team
- Report to a Senior Director of Product Design
- Key calls
- AI as the default starting point, not a side experiment
- Designers direct, edit, and validate instead of producing from scratch
- Build the tools the team runs on, then hand them over
The short version, if you would rather watch than read.
Watch the short version · 55 seconds
01The team is the point.
Team structure
Coach the thinking, not the pixels
I coached a designer to start with the problem before visual polish. We stripped out color to focus on the fundamentals. She then ran first-round interviews for a senior hire and led our color filter work from analytics data.
Room to grow
Growth here means more ownership. Designers run customer interviews alongside our UX researcher, have a say in our hiring, and lead our biggest projects. One designer moved into running a major project as its product manager. My senior UX lead manages a junior designer, so leadership is spread across the team instead of stuck with me.
Fix the handoff
An under-scoped handoff cost the team extra design cycles. I built a more rigorous intake process so design asks its questions up front.
Keep the team at 80%
I plan about 32 of each person’s 40 hours a week for project work. The other eight cover meetings, urgent asks, and learning, so nobody has to work late to keep up. Incubator projects live inside those 32 hours, not on top of them, and a dashboard flags anyone running over. I would rather the team be consistently reliable than occasionally extraordinary.
Prototypes, not wireframes
My UX designers now hand off working prototypes with real interactive states, not wireframes. The UI designer then fits them to the design system and leads with engineering. When two fully coded prototypes went to developers, the receiving engineer said no additional work was needed. Most designers see their own projects through to launch.
When I build, and when I delegate
My designers own consequential work, including the long, ambiguous project in the next section, and I manage capacity through my lead. I step into select technical builds and urgent prototypes myself, but with a bounded scope and a handover. The MCP is one example: I built and demoed the V1, and it is in pilot now, with the production and authentication handoff to IT still in progress. The point is not to keep the hard work. It is to set the bar, then hand it over.
02The calls I made.
Build it or buy it
- The problem
- The research team wanted a faster way to get answers out of its studies. A vendor quoted a per-seat yearly fee for the one feature we needed, bundled with a dozen we did not. With over 100 people using our research platform, that was a six-figure bill every year.
- The options
- Sign the contract, or build that one feature ourselves.
- My call
- Build it. I built and tested it the same afternoon the proposal came in, so there was nothing left to debate.
- Who carried it
- I built it. It connects to our research library and answers questions in plain language. Then I turned it into a Teams assistant, so anyone can look up our research on their own.
- Where it landed
- About 28 people across the digital team get answers in minutes instead of digging for hours, and we never signed the contract.
- Looking back
- I would bring IT in on day one. The build took an afternoon, but the handoff to run it in production is still in progress.
Learn the tool before the ticket
- The problem
- The business picked a new search platform for the whole site. It would hit hardest on our listing and product pages, where most traffic flows, and my team had nearly finished redesigning the product page.
- The options
- Wait for the integration to be handed to us, or learn the tool now and design for it before anyone asked.
- My call
- Do not wait. I asked the whole team to learn what the new search tool could and could not do before the work reached us.
- Who carried it
- The team. They designed how search would look and work on our site, then worked it through with our IT and enterprise partners.
- Where it landed
- It is live. Most of the redesign carried over, because search changed what loads on the page, not the design underneath it.
- Looking back
- I would ask for a seat in the search decision itself, not just get ready for it, since it shapes our busiest pages.
Hand the hard problem to the right person
- The problem
- A foundational project had sat with another team for two years or more. It changes how the platform works, not just how customers use it.
- The options
- Hand it to an agency, or bring it in house and work it through with our own partners.
- My call
- In house. An agency would have been slower and cost more, and nobody knows the business like the people inside it. It also let us build stronger ties with IT and sales.
- Who carried it
- A product and UX designer I brought in from my network, the most technical person on the team and great with stakeholders. He mapped what the business could support, then led the design. We turned it around in a few months.
- Where it landed
- A steering committee narrowed it to one focused problem, customers tested the prototypes, and engineering is building it now. A change this deep takes a lot of sign-off, so it has not shipped yet.
- Looking back
- I would push for that narrow scope at the start, instead of waiting for the steering committee to get us there.
Build the fix customers kept asking for
- The problem
- Customers kept asking for a way to cancel an order right after placing it. It came up nearly 50 times across two years of our research. We had one day at a company hackathon to fix it.
- The options
- Other sites give customers somewhere between 15 and 30 minutes to cancel. We had to pick a window.
- My call
- 30 minutes, the high end of the range, so nobody runs out of time to fix a mistake.
- Who carried it
- A team of five. I pulled the idea from our research and designed it, and our lead engineer built it in about half a day.
- Where it landed
- We won. The win led to an immediate push to build it for real and to run hackathons more often. It is fully built and waiting on release, and its second phase is already designed.
- Looking back
- I would line up the release path before the hackathon, so a winning idea does not sit waiting to ship.
Map it before you draw it
- The problem
- We were about to start the screens for a new notification center, which has to handle many different alert scenarios.
- The options
- Start the UI right away, or map a decision tree for every scenario first.
- My call
- Map it first, so the screens answer settled questions instead of open ones.
- Who carried it
- A designer on my team leads the design. We mapped the trees together, then walked them through with product and engineering.
- Where it landed
- The stakeholders aligned on the trees, the delivery plan is locked with ticket dates, and the mobile prototype is moving.
03I build the leverage.
How the team shapes the roadmap
I founded the research practice and built the discovery framework the team runs on: the customer problem, the market opportunity, the alternatives, and how we will know it worked. Every designer carries at least one incubator project next to their roadmap work, and most of the roadmap starts as work we incubate and validate this way, and one opportunity the team surfaced became a top priority for an executive the next year.
Stay close where the stakes are highest
The product page is where millions of dollars in revenue move every day. A senior designer, our researcher, and I tested the redesign with customers and kept changing it based on what they told us, like a faster inventory drawer and color changes that load in under a second. It is live, and the next version is already in design. Looking back, I would set shared UX review points earlier on a page this important, so fewer changes land late.
One design system, not three
My team is rebuilding a ten-year-old patchwork of three agency-built frameworks into one design system, with shared names for color, spacing, and type, a curated icon set, and a staged rollout. The latest version is live in Figma, and the docs show which parts are already coded. We also built a browser extension that checks live pages against the system and flags mismatches and accessibility issues, so drift gets caught instead of piling up.
Shipped, and used
It is not all internal. A catalog builder I led shipped to the platform in 2025. Customers and staff have used it to make thousands of product catalogs to pitch to their own clients.
The real screens live here
Blurred on purpose. Ask me and I will walk you through the real thing.
Ask me for the walkthrough →Shipped and used
Thousands
of customers and staff use a design tool I shipped to the platform, and it has produced thousands of designs.
Product analytics · Adoption, not conversion
04The slow parts, removed.
One platform, not tens of thousands of storefronts
One shared platform is the goal, and the company’s first MCP is the step toward it. I designed and built it so AI tools and platforms can pull live product data through one integration instead of one per platform, with the brand and pricing rules built into that layer so they travel with the data. I built and demoed a working V1 to our AI governance leadership, and it is in pilot now. It is the foundation for serving from one platform what used to take tens of thousands of custom storefronts. If we are talking, ask me for the walkthrough.
The real screens live here
Blurred on purpose. Ask me and I will walk you through the real thing.
Ask me for the walkthrough →Less labor on launch-catalog writing
Getting a wave of new products live used to mean two writers spending a full month on the launch-catalog copy. An agent I built now does that writing in under an hour, and it lands as a one-click import. The team gets those weeks back for the work that actually needs a person.
Before: writing labor
A month
Two writers on a launch-catalog writing task.
After: agent task
<1hour
An agent I built runs the writing task.
My own before and after · Writing labor, not launch time
05From novelty to default.
What I drive
I drive AI adoption across teams, working with our AI enablement team: a weekly governance cadence, operating rules, executive alignment, an AI roadmap, and a listening tour that shaped the team adoption model.
Figma adoption
2 → 300+
Figma users across the org, from a couple when I arrived to more than 300.
Protected case study
The full screens and process for this work stay private out of respect for the company. If we’re talking, ask me for the password, or reach out and I’ll walk you through everything.
That’s not it. Ask me for the current password.
Want the story without the password?
