The UX process end to end
A lightweight path from a fuzzy problem to a shipped feature: discover, define, develop, deliver — diverging then converging, twice.
In one line: good process is a rhythm of opening up to explore and narrowing down to decide — done deliberately, it turns a vague brief into something worth building and then actually builds it.
What it is
The UX process is the sequence of work that takes you from "we think there's a problem here" to "we shipped something that helps." A useful way to picture it is the double diamond: two diamonds side by side. The first is about finding the right problem (discover, then define); the second is about finding the right solution (develop, then deliver). Each diamond widens — you generate options — then narrows — you commit.
The widening is divergence: explore broadly, gather signals, sketch many directions, resist deciding. The narrowing is convergence: synthesize, choose, refine. The skill is knowing which mode you're in, and not collapsing them — deciding while you should still be exploring is how teams build the wrong thing quickly.
Why it matters
Skip discovery and you solve a problem nobody has. Skip definition and you boil the ocean. Skip the deliberate divergence in develop and you ship the first idea, not the best one. The process isn't ceremony — each phase exists to answer one question and produce one artifact, then hand off a clearer thing to the next. Sized to a two-week project it's light; sized to a year it scales. What stays constant is the rhythm: open, then close; open, then close.
See it
Tweak it3
How it works
Each phase answers a question and leaves something behind:
- Discover (diverge). What's really going on? Talk to people, watch them work, gather signals. Output: a pile of honest observations.
- Define (converge). Which problem is worth solving? Synthesize the pile into one sharp problem statement. Output: a focused brief.
- Develop (diverge). What could we build? Generate many solutions — sketches, flows, prototypes. Output: candidate directions.
- Deliver (converge). What actually works? Test, narrow, refine, and ship. Output: a real thing in front of real users.
You can loop back at any boundary — discovery in deliver often sends you to redefine. The diamonds are a rhythm, not a one-way street.
Build it
A phase tracker is a small but real piece of UX: it tells someone where they are and what's next. Here's an accessible one that marks the current phase with aria-current, so the meaning reaches assistive tech, not just sighted users.
<nav aria-label="Project phase" class="phases">
<ol class="phases__list">
<li class="phases__step">Discover</li>
<li class="phases__step" aria-current="step">Define</li>
<li class="phases__step">Develop</li>
<li class="phases__step">Deliver</li>
</ol>
</nav>.phases__list {
display: flex;
gap: 0.25rem;
list-style: none;
margin: 0;
padding: 0;
}
.phases__step {
flex: 1;
padding: 0.5rem 0.75rem;
text-align: center;
font-size: 0.85rem;
border-radius: 0.5rem;
background: #eceef3;
color: #5b606b;
}
/* The current step is bolder AND outlined, so it reads without color. */
.phases__step[aria-current="step"] {
background: #1f2330;
color: #fff;
font-weight: 700;
outline: 2px solid #1f2330;
outline-offset: 1px;
}
@media (prefers-reduced-motion: reduce) {
.phases__step {
transition: none;
}
}Make it yours
Use the controls beside the demo above to change phase, show diverge / converge labels, and show problem / solution space — each change updates the example live.
Experiment in the playgroundReproduce it with an LLM
Reproduce it with an LLM
You are a design lead scoping a two-week project for a small team. Lay out a lightweight UX process for building '{a budgeting feature for a banking app}' from discovery to delivery. For each phase (discover, define, ideate, prototype, test, ship), give: the question that phase answers, one or two activities sized for two weeks, the artifact it produces, and the signal that you're ready to move on. Avoid ceremony for its own sake. Return a markdown list grouped by phase.
Pitfalls & accessibility
Related
Design thinking
A five-mode loop — empathize, define, ideate, prototype, test — for turning a fuzzy human problem into something worth building.
Customer journey mapping
Map a goal across stages — doing, thinking, feeling — to find the friction moments and the opportunities hiding in them.
User research basics
Match the question to the method: what people say vs what they do, exploring vs evaluating — and how many participants is enough.
Further reading
- A primary write-up of the double-diamond framing from a national design body, in your own words.
- Any solid text on lean, timeboxed product discovery, for how to size each phase to the work.