The 5-Day Validation Sprint Playbook

The exact agenda, prompts, and decision artifacts I use to compress three months of pre-build research into five working days. Tested with Parliament, Admicom, and the agency partners I work with.

• The 5-day agenda, hour by hour
• The pre-sprint memo template (so the team aligns before the kickoff)
• AI-augmented prototype stack: tools, prompts, and where each one breaks
• User recruit script (the version that gets you 5 in 24 hours)
• Decision artifact template — the one-pager engineering will actually use
• How agencies frame the sprint to their clients without giving away methodology

1. Day 1

   Frame the bet

   • Translate the business decision into one user-facing question worth testing.
   • Lock the target user, the decision criteria, and the kill switch.
   • Pre-write the post-sprint memo: what will we believe by Friday, and what would change our mind?
2. Day 2

   Map what's known

   • Two-hour expert sessions with internal stakeholders. No slides — only questions.
   • Surface every assumption that, if wrong, kills the bet.
   • Rank assumptions by risk × reversibility. Top three become the prototype targets.
3. Day 3

   Build the smallest believable prototype

   • AI-augmented production: Figma + Make + v0 + Claude Code, scoped to a single user task.
   • Cut anything that doesn't directly test the top-three assumptions.
   • Recruit 5 users from the actual target segment for tomorrow.
4. Day 4

   Test with real humans

   • 5 sessions, 45 minutes each. Same script, same prototype, different brains.
   • Capture verbatim quotes against each assumption, not against the UI.
   • End-of-day debrief: which assumptions held, which broke, which need a second look.
5. Day 5

   Decide and hand off

   • Fill in the pre-written memo with evidence.
   • One-page recommendation: build, don't build, or test further.
   • Hand off the prototype + recordings + memo to engineering, leadership, and (if applicable) the agency partner.