Ask for Outcomes, Not Components
Ask for Outcomes, Not Components
Section titled “Ask for Outcomes, Not Components”The single most important rule in this course.
The rule
Section titled “The rule”When you prompt Claude Code, describe the user outcome you want, not the component you think will deliver it.
Components are conclusions. Outcomes are problems. Claude is much better at picking the right component when it understands the problem.
Three pairs
Section titled “Three pairs”Pair 1
Section titled “Pair 1”Weak prompt:
Add an accordion.
Better prompt:
This settings page feels too dense. Recommend whether accordion sections, tabs, a right-side drawer, or progressive disclosure would make it easier to use. Explain the tradeoffs, then implement the best option.
The first asks for a thing. The second asks for an outcome.
Pair 2
Section titled “Pair 2”Weak prompt:
Add a menu.
Better prompt:
This page has too many visible actions. Recommend the best action organization pattern: dropdown menu, toolbar, split button, context menu, or command palette. Explain which actions should stay visible and which should move into the menu.
The first asks for a component. The second asks for behavior change and lists rivals.
Pair 3
Section titled “Pair 3”Weak prompt:
Make the dashboard better.
Better prompt:
Improve this dashboard for a user who needs to understand performance quickly and take action. Focus on metric hierarchy, filters, drill-downs, table actions, empty states, loading states, and responsive layout. Give me 3 design options before coding.
The first asks for “better.” The second asks for “better for whom doing what.”
Why this rule beats every shortcut
Section titled “Why this rule beats every shortcut”- Specificity beats generality. “Better” is unmeasurable. “Easier to scan in 3 seconds” is.
- User-first beats component-first. The right component changes when the user changes. Lock in the user, let the component follow.
- Rivals beat singletons. Naming three rival patterns forces Claude to pick — and to defend the pick.
- Tradeoffs beat assertions. “Implement the best option” without “explain the tradeoffs” lets Claude stop thinking. The phrase forces analysis.
A four-part formula
Section titled “A four-part formula”Every UI prompt fits this template:
[The user problem in 1 sentence].
Recommend the best pattern from: [3-5 rival patterns by name].
Explain the tradeoffs, then implement the best option.
Verify with [typecheck / lint / responsive / a11y].Memorize this template. Run it on every UI request. After two weeks, you’ll do it without thinking.
What this rule unlocks
Section titled “What this rule unlocks”- Compound vocabulary. Every prompt teaches you 2-3 new pattern names.
- Compound reuse. Every “rivals” list reminds Claude what your stack has.
- Compound discipline. Every “explain the tradeoffs” stops Claude from skipping straight to code.
- Compound trust. Every “verify with” stops “done” from meaning “compiled.”
Four compounds × every UI session × every day = 100x.
That’s the rule. That’s the course.