Product
Designer

Missions are opt-in tasks a user takes on to manage a chronic condition. A bit of reading. A daily walk. A moment of reflection. Nobody had built quite this before, and we needed around six hundred of them, diverse enough to stay interesting across a wide range of goals.

That number is the whole problem. At six hundred, every mission being its own bespoke thing is an unmaintainable mess. So the real question was never what makes one good mission. It was: what’s the smallest set of repeatable parts that can generate six hundred distinct-feeling missions without six hundred distinct builds?

Built the system from the floor up rather than imposing a taxonomy from above, working with our UX researcher and asking the same question over and over: what is the most basic mission that can exist? The simplest is reading plus illustration. Then an out-of-app behavioral task: park farther from work, take the extra walk, five days running. Growing the grammar outward from the simplest real units is more honest and more durable than starting from abstract categories.

One early decision did a lot of quiet work: missions run a fixed one to five days. Not because it was tidy. Every extra variable is one more thing a user has to be taught. Constraining the system protects their attention, not just the design.

Every mission shares one three-act spine: preparation, action core, reflection. Each act holds a kit of swappable modules, and modules appear only when the mission needs them. A walk needs step-tracking and a reminder; a reading mission needs neither. A mission is assembled, not drawn from scratch.

Today the library runs about fifty-five modules across the three acts, and a new mission is a set of module choices rather than a new design. The reflection act closes the loop. Users rate a mission on a three-point scale, and that rating feeds what gets recommended next, so the system teaches itself as it runs. Future module types like AI chat, medication tracking and mood logging drop into the right act without disturbing the spine.

Honest scope: I designed the system and its module library. This work is current and unreleased, so what I can tell you is true about the structure and the decisions, not about how it performs in the world. Happy to walk through the full case live.

In a shared story anyone can attempt anything, including things the world can’t allow. The question was never whether to block the action. It was how to deny it without breaking immersion or the player’s trust in front of everyone else at the table.

Explored three directions and took a position. A lightweight block was the most feasible: it meets the safety need but hides the why and reads cold. A private message was the future state. I recommended the in-character refusal. The narrator denies the action in-voice, strikes the impossible intent, and redirects with humor, keeping the moment inside the story instead of breaking it.

I presented the range and the tradeoffs to the founders. We shipped the lightest slice, and I documented the richer in-character and private-message versions as the path forward.

I left before seeing it implemented, so this study is about the problem, the exploration and the decision rather than a measured outcome.

The product’s initial visual direction came from an agency. Leadership loved it. It arrived without a style guide.

Designer after designer tried to extend it and came back with notes. The work kept missing something nobody could name. Illustration became a point of contention in design reviews. Even when images were labelled as placeholders, leadership would zero in on them instead of the flows.

I was tasked with designing a novel part of the experience: Missions. Small, doable activities, each a step toward a user’s health goal. They’re heavy with illustration and carried a lot of the agency’s original influence. So I studied what the agency delivered and what other designers had attempted, to understand what leadership was reaching for and what it was rejecting.

Photographic cutouts, singular and clearly silhouetted, on a neutral ground. One to three objects with room to breathe. Graphic accents like motion lines and sparkles that add energy without dominating. One consistent lighting direction, soft, no harsh contrast. Natural tones with pops of bright color, never deep saturation or muddy hues.

The harder half was tone, because that was the part getting picked apart. I wrote it as boundaries rather than adjectives: playful, not childish. Insightful, not academic. Warm, not saccharine. Conceptually clever, not punny for pun’s sake. Grounded in realism, but with an imaginative twist. Every rule names the edge it stops at, which is the thing an adjective on its own can’t do.

Then the method: read the mission, find the emotional core, brainstorm objects that carry it, add one twist. A real goose beside a decoy for a mission called Fact or Fiction. Capture the essence, not the action.

Eight months as the sole illustrator left a large library of illustrations, elements and motion the team could build on, then a team of three producing to the standard. Designers pull from the repository and feel safe in reviews. The tone has become legible to people who have never opened the style guide.

Leadership can now identify my work without being told. They call it out, and they ask me to carry that quality across the app, to find ways to bridge rougher moments with motion, illustration, or interaction.