Overview

This template provides a structured approach to running a 90-day AI pilot programme. It's designed for mid-market companies deploying their first or second AI initiative and includes everything you need to plan, execute, and evaluate the pilot.

Phase 1: Design (Days 1–14)

Define the Pilot Scope

Scope Template

Business problem: [Specific, measurable problem statement]
Target outcome: [Quantified improvement expected]
Department/team: [Pilot participants]
AI solution: [Tool or system being tested]
Budget: [Total pilot budget including tools, time, and training]

Success Criteria

Define 3–5 measurable criteria. Example:

Reduce [process] time by [X]%
Maintain quality score above [threshold]
Achieve [X]% user adoption among pilot participants
Stay within [budget] for the pilot period

Decision Gates

Critical Checkpoints

Day 14 — Go/No-Go: Technical setup complete, team trained, baseline metrics captured
Day 45 — Mid-Point Review: Early results reviewed, adjustments made, blockers escalated
Day 90 — Final Evaluation: Full results assessed, scale/pivot/stop decision made

Phase 2: Execute (Days 15–75)

Weekly Cadence

Monday: 15-minute stand-up — blockers, priorities, quick wins
Wednesday: Data collection checkpoint
Friday: Week-in-review — metrics update, learnings captured

Data Collection Template

Track weekly:

Usage metrics (frequency, duration, features used)
Performance metrics (against success criteria)
User feedback (satisfaction score + qualitative notes)
Issues and incidents log
Cost tracking (actual vs. budgeted)

Phase 3: Evaluate (Days 76–90)

Final Report Structure

Executive summary (1 page)
Results vs. success criteria
Financial analysis (actual ROI vs. projected)
User feedback summary
Technical assessment
Recommendation: Scale / Pivot / Stop
If scaling: proposed timeline, budget, and resource plan

Appendix: Risk Register Template

For each identified risk, document: description, likelihood (H/M/L), impact (H/M/L), mitigation strategy, and owner.

90-Day AI Pilot Programme Template