AccessWear: Designing a Circular Outerwear Service

AccessWear is a circular outerwear subscription: $50/month for seasonal access to premium jackets across 10+ brands. I joined two weeks after launch, when the platform had 200 people waitlisted and fewer than 5 paying. Over nine months our team rebuilt the service end to end and the growth system behind it, taking it to 120+ members and a 10/10 NPS, and proving people wanted it.

Our team at the consultancy worked with the client's innovation team to test a new model: recurring access instead of one-time, $600–800 jacket purchases. We ran it as a Wizard of Oz pilot, operated by hand the whole way, so we could test real demand before building anything to scale.

This was a team effort at our consultancy, built with the client's innovation group, so the rest of this story is told as we. Two parts of it were mine end to end.

UX design: I redesigned the homepage, browse, product page, checkout, How It Works, and About. Supply chain and data: I ran fulfillment, procurement, and order management by hand, and I built the tracking that gave the team data across the whole pilot, from ad performance to swaps.

Early usability testing found two main problems: no trust signals (no social proof, no founder story, no reviews) and a confusing path from browsing to subscribing. People weren’t rejecting the idea of a subscription. They lost confidence partway through and left.

The numbers backed it up: a 4.7% waitlist-to-subscriber rate against a 10% target, and 65% of homepage visitors leaving above the fold. The platform ran on a patchwork of Shopify, Stripe, and a separate customer portal, with broken handoffs in checkout.

Sixteen moderated usability sessions in the first month surfaced three problems, and they showed up across every type of user.

First, no trust. The site had no social proof, no reviews, and no clear voice. People called it anonymous and wanted to know who was behind it before committing to a monthly charge.

Second, the path to subscribe was broken. There was no way to filter by activity, and checkout ran across three disconnected systems. People lost their place and left instead of pushing through.

Third, seasonality, and it ran deeper than timing. A late-spring launch meant we were selling winter shells into summer. So we used the slow months as setup time: fix the experience and build the audience, then be ready when cold-weather demand arrived.

The team ran lean, with no dedicated engineering. The supply chain was mine: every order meant procurement, inspection, packing, and shipping by hand, and I built the order-tracking and inventory system from scratch so we had clean data across the pilot. The design and front-end changes were mine to ship too.

The catalog spanned 10+ outdoor brand partners, each with its own product cycles, asset specs, and approvals. Running without a real fulfillment system was a deliberate cost choice for a pilot.

We split the work into three phases, timed to the season rather than to a sprint calendar.

Phase 1 (Stabilize, May to July): fix the broken funnel, add trust signals, and run cheap acquisition tests to see which channels and messages were worth more.

Phase 2 (Build, August to September): with the core flow stable, test messages across paid and organic, and build the email flows that keep members before peak season.

Phase 3 (Convert, October to December): turn it all on as demand rose, with brand co-marketing, a community launch, and steady outreach to warm leads.

The first sprint went after the homepage and checkout. I rebuilt the top of the homepage to fix the 65% drop-off and restructured browse with activity filters, while we fixed the Shopify and Stripe handoffs that were breaking checkout. A lead-capture popup pulled 15 sign-ups in its first weekend.

To close the trust gap, I pulled product reviews from brand partners and added a founder story. Sign-ups climbed from 9 to 17 in the first cycle. A second round of 16 sessions surfaced the next barrier: people couldn’t read the value until they saw the subscription price next to the jacket’s retail price.

By the end of the pilot the service was one clean experience instead of a patchwork. Here is the live build, one feature at a time, each one interactive.

With the funnel stable, we moved to messaging. We ran 18 ads, each testing one value proposition. Try-before-you-buy won, with click-through up to 6.56% and cost per click as low as $0.24.

We tested channels beyond ads too, in-person activations, run-club and outdoor-media partnerships, and a Canadian launch, and built a 14-step email flow for retention. By the end of the phase the pilot had earned an extension, the sign the bet was working.

October was the shift from awareness to conversion. We ran co-marketing posts with six brand partners, each one adding followers. Outreach to 500+ accounts a week opened 1,500+ conversations at a 10% reply rate, and we went back to people who had already followed or commented, pulling about 300 of them toward the Insider Program.

We launched the AccessWear Insider Program on Discord, a community where the most active members earned access to $700–800 jackets. It became the best feedback channel we had: members shaped the product and stuck around because they felt part of it.

Growth built through the quarter. The Instagram account climbed from 3,000 followers in mid-October to nearly 9,000 by late November, and that audience began to convert, monthly active members more than doubled over the same weeks. By December 120+ people had joined. We held active membership at 85–100 at a time, the limit of what I could fulfill by hand, not what people wanted.

The pilot proved the core idea: premium outdoor buyers will pay to access jackets on a subscription when the experience is smooth and the catalog is worth it. Members gave a 10/10 NPS, and what they did on their own, posting trip photos, bringing in friends, asking for new gear categories, told us more than the headline numbers. About one in ten members swapped jackets, on average a month apart, the rotation the model was built for.

Churn data deepened the picture. The primary drivers for leaving were seasonal timing (80%) and limited jacket variety (20%), constraints of when we launched and what was in stock, not dissatisfaction with the model. When asked what would bring them back, most pointed to better year-round and seasonal relevance, with broader selection close behind. Both are addressable: seasonal fit through launch cadence and a year-round catalog, variety through scale.

Brand partners leaned in regardless of subscriber count: some moved to joint marketing arrangements, others co-ran giveaways, and several of their leaders asked for follow-ups. A co-founder of a major rental-fashion company advised on the financial model.

The email flows we built in September should have been live in July; the delay cost conversions in the months when trust mattered most. The Supercycle integration that would have fixed checkout sat too long on the back burner. The Discord community, the best feedback channel in the whole project, should have been there from day one instead of arriving in month four.

The pattern I took from all three: in early-stage work, retention and feedback almost always need to come before acquisition. Build the loop first, then scale the funnel. The other way around costs you.