Where's my rug?

The data was there. Just not where customers expected it.

Ruggable had two post-purchase pages: an Order Details page (where customers landed) and an Order Status page that lived behind a "Track Item" link inside Order Details. The status data customers actually wanted (shipping state, tracking, delivery estimate) was on the status page, one click further than most people went.

WISMO ("where is my order?") was 10% of inbound CX tickets, the largest single category by volume. CX brought the issue to UX with a clear ask: consolidate the two pages into one, redesign around what customers actually need on the page, and deprecate the Order Status page.

01 Order Details where customers landed

02 Order Status one click further

How a project moves at Ruggable.

Tickets at Ruggable start with a PM, who scopes the problem and writes the brief. UX takes it from there. Research, references, ideation, then rough directions back for review. We loop with PM and the rest of UX first, then bring CX in. Engineering builds in continuous feedback rounds with us so what ships looks and reads right.

For this one, CX had pulled both the volume picture and the qualitative. My job wasn't to gather raw research. It was to turn the brief into design directions.

Reference work first

Before opening an artboard, I built a reference board in Figma. Post-purchase tracking patterns from peer DTC brands, carrier-side experiences, app and email order updates, and the existing Ruggable pages annotated with what was working and what wasn't. The reference board became the team's vocabulary, so when we discussed concepts later, we could point at it.

The configuration space, not the directions.

The reference board pointed at a card system early, so the interesting question wasn't which approach. It was which configuration: what goes where, in what order, at what hierarchy. I worked through that exploration on both surfaces, with the same cards composed differently each time.

Mobile · top-area variants

On mobile, the variation lived almost entirely in the top section. The answer to "where's my rug?" had to come first, but how to compose ETA, status, and the product itself was open. Below the fold the content was identical across variants.

Desktop · full-page variants

Desktop opened up at the page level. The variation moved to multi-item orders, where to slot promotional touts, and how to surface the latest update without burying the timeline. Each variant tested a different balance.

The resolved direction

We landed on the expand-and-collapse pattern. Each item card minimizes to a compact row by default and expands for the full tracking detail. It handled the multi-item case cleanly, kept simple single-item orders from feeling sparse, and gave touts and updates a stable home that didn't shift between states.

Listed every card before building one.

Before drawing any UI, I listed every card the system would need: order status, ETA, tracking number, line items, payment summary, customer service link, delivery address, exception messages, line-item drilldown, shoppable upsell. The list kept me organized as I worked. I could see what each configuration was missing, drop in new cards when ideas came up, and decide which earned a spot in the final set. MVP scoping became a content decision rather than a layout one.

I designed each card individually (default, empty, edge cases), then assembled layouts from them. Each state got its own configuration: order received, label created, shipped, out for delivery, delivered, delayed, returned, plus the logged-in / logged-out variants and a handful of error states. Some cards earned a permanent slot. Others got pulled into MVP-deferred (line-item drilldown, shoppable elements, how-to LP links) so we could ship the most important cards first and add the rest later without redesigning the page.

Team review & CX feedback

Two or three solid configurations went to the team, PM and my UX lead first, then CX. CX feedback was mostly content-level: phrasing, hierarchy, copy that read like a person rather than a system. Because CX had been in the loop as observers from the early rounds, the formal review was a tightening pass rather than a reset.

One order can be many shipments.

A Ruggable order isn't always one box. Items get grouped onto shipping labels and bundled into the parcels they actually travel in, so a customer who ordered several rugs can end up with three or four shipments, each one moving on its own schedule. The redesigned tracker stacks each shipment as its own card: a plain-language status up top, expandable to the full timeline and the items inside it when the customer wants the detail.

We designed for the worst case on purpose. The old tracker assumed the simple order, one box, one timeline, and fell apart the moment a customer ordered a lot. Building the stacked, expandable pattern around the messy order first meant the simple order was just the one-card version of the same system, not a separate design.

Language that calms the wait

The biggest copy work happened in the in-between states, the days after a customer submits their order but before the package physically ships. That gap is where anxiety builds: "is my order actually being made? did anyone see it? when does this turn into a tracking number?" The old page just said "Order Received" and went quiet.

The new copy answers the unasked question. Each state surfaces an expectation the customer would otherwise email about, things like "typically ships within 1 week" or "scheduled to ship by Jan 8", paired with the status itself. Same data the system always had; we just decided to say it out loud instead of leaving it implied.

A true mobile experience, not a port.

Per the data team's analytics, a significant share of order-tracker traffic was mobile. I didn't design mobile first, but I treated it as a real surface. Single-column, generous tap targets, one primary action per state, and content prioritized for the smaller view. Mobile got the same care as desktop, not a shrunk-down version of it.

01Order received

02Out for delivery

03Drawer expanded

The ticket pile shrunk.

After ship, our PM tracked WISMO ticket volume and ticket types against the old experience. Within 90 days, the result was clear.

Post-ship "where is my package" inquiries were effectively eliminated; the remaining ones are edge cases the page deliberately escalates. CX redirected the saved hours into proactive damage-claim outreach.

What I'd do differently.