Brief
I set this brief to close a gap in my own portfolio: everything else I'd built was a single-purpose site. I wanted to design a full loop instead, discover, save, organize, to show I can think about information architecture and flow, not just one polished screen.
Who it's for
People planning a room, a move or a renovation who collect ideas over weeks, not minutes. My working assumption, still untested, is that their inspiration ends up scattered across screenshots, saved posts and open tabs, so the hard part isn't finding ideas. It's finding them again when it's time to decide.
Process
I mapped the core loop first: browse a feed, save something in one tap, find it again later in an organized board. Every screen after that was built to protect that loop rather than add features on top of it.
Key decisions
- A masonry grid on the discovery feed, since inspiration content is naturally different aspect ratios, and forcing it into uniform tiles would flatten exactly what makes it feel curated.
- Save-to-board as a single tap on the feed itself rather than a separate page, since adding friction to the core action of the whole app was the one thing I didn't want to trade away for polish.
- A serif display face paired with a warm, muted color system to make the app feel like an edited collection, not a utility, deliberately different from Weekflow's quieter, more functional register.
Scope
Anything that didn't strengthen discover, save and organize stayed out of the first version.
- In: onboarding, the discovery feed, item detail and boards. Together they're the smallest set of screens that completes the loop.
- Out, for now: following other people, comments and shopping links. Each could earn its place later, but each adds a reason to leave the loop before it's proven that people come back to it.
Measuring success
Palette is a concept, so these are the measures I'd use if it were built. All of them test whether the loop holds.
- Activation: a new user saves at least one item in their first session. Onboarding's style picker exists to make that first save easy, so the feed isn't empty and generic.
- Primary: people come back and open a board within a week of saving to it. Saving is easy; returning to what you saved is the real test of the product.
- Guardrail: drop-off at the style picker. If asking for preferences up front loses people, it's costing more than the better feed is worth.
Before any build, I'd test a clickable prototype with five people planning a real space and watch whether they can save something and find it again unprompted.
Outcome
Four core screens designed end to end: onboarding, discovery feed, item detail, and board view, as a coherent system sharing one navigation pattern and one visual language.
Honest note
This is a concept exploration, not a built or coded product, and the imagery is a curated color system standing in for photography rather than real photos. I designed it this way deliberately, to keep the focus on layout, hierarchy, and flow rather than sourcing.