Validation

You are helping an Innovation Accelerator design a 30-minute moderated interview to test the desirability of an early prototype.

Context:
- Problem we think we're solving: [problem]
- Target user (role, context, current workflow): [user]
- What the prototype does today: [prototype summary]
- The riskiest desirability assumption: [assumption]

Produce:
1. Three behavioral warm-up questions about how they handle [problem] today — no hypotheticals.
2. Five open questions that surface pain, workarounds, frequency, and cost of the current approach.
3. Three "reaction" questions to ask AFTER they've used the prototype, focused on what they'd change in their workflow, not whether they "liked" it.
4. Two commitment probes (would they pilot it, share it, give up something they use today).

Avoid: leading questions, "would you use this", "wouldn't it be cool if", and anything that asks the user to predict their own future behavior. Flag any question that violates this and rewrite it.

Help me design an unmoderated usability test in Maze for the prototype below. Target 8-12 participants, ~15 minutes total.

Prototype: [link or short description of the flows available]
Primary user goal: [goal]
Success criteria I care about: [e.g. completes checkout without help, finds the report in <60s]

Produce:
1. A 2-sentence intro to show participants (no product marketing, no leading language).
2. 4-6 task prompts written as user goals ("You need to…"), not instructions ("Click the blue button"). Each task should map to a success criterion.
3. For each task: one post-task follow-up (open text) and one 1-5 confidence rating.
4. Three post-test questions: one about what was confusing, one about what was missing, one behavioral (what would they do next in real life).
5. A screener with 3 questions to filter for [audience].

Call out any task where success is ambiguous or where the prototype likely can't support the path.

You are a pragmatic staff engineer reviewing an early prototype for feasibility risk before we invest in a production build.

Prototype summary: [what it does]
Data it touches: [systems, PII, volumes]
External dependencies: [APIs, models, vendors]
Constraints we already know: [SLA, latency, compliance]

Produce:
1. Top 5 feasibility risks, each with: risk, why it matters, cheapest experiment to de-risk it in <2 weeks.
2. Any assumption in the prototype that would break at 10x or 100x today's volume.
3. Data / integration questions we should answer BEFORE handoff, phrased so a delivery team could actually answer them.
4. A short "won't work as-is" list — parts of the prototype that fake something the real system can't do that way.

Be blunt. Do not hedge. If something is fine, say fine.

Help me pressure-test the viability of this opportunity before we recommend it move to delivery.

Opportunity: [1-2 sentence description]
Who benefits and how: [users, business unit]
Rough cost to build & run (best guess): [cost]
What we'd have to stop doing to fund it: [tradeoff]

Produce:
1. Five questions I should be able to answer for a sponsor before asking for funding — organized as: value, cost, risk, timing, alternative.
2. The two most likely reasons a CFO or portfolio owner would kill this, and what evidence would defuse each.
3. A "smallest defensible pilot" — the narrowest slice we could run to prove the business case is real, with a rough success metric.
4. One question I'm probably avoiding because I don't want to hear the answer.