a one-paragraph summary of what this was, who it was for, and what changed because of it. this is the elevator pitch — the single thing someone scanning the page should walk away with.
every project starts with a problem. describe the world before you got involved — who the users were, what they were trying to do, where the existing experience was breaking down, and why this mattered to the business.
keep it human. specific. avoid the corporate-deck voice. "merchants kept calling support to ask the same three questions" beats "we identified opportunities to optimize self-service."
a short prose section explaining how you tackled it. what you ruled out and why. what hypothesis you went in with. the constraints you were working under (timeline, headcount, technical debt, that one engineer who really hates radio buttons).
interviews, support tickets, session replays. find the actual problem under the reported one.
one sentence: "if we do X, we believe Y will happen, and we'll know because Z."
what's the smallest thing we can put in front of users in 3 weeks that proves or kills the bet?
watch the metrics. ship the next slice. or kill it gracefully and write the postmortem.
a comparison block. doesn't need a real screenshot — describe the difference plainly, and the placeholder mock below stands in until you have the real visuals.
describe what the user encountered before — the friction, the steps, the error states.
describe the new flow, the new affordances, the gnarly edge case you finally solved.
what actually happened? numbers if you have them, qualitative wins if you don't. be honest about what didn't work — the page reads more credibly when you say so.
the section everyone skips on portfolios. don't skip it. one or two real reflections — what you got wrong, what you'd run faster next time — tell readers more about you than the wins do.