Professional Validation Library

The AI Product Manager's Playbook · Kai London

Discovery, Evals, Pricing and Launch for AI Features

Twelve fabrications in 3,920 items look like 0.31 per cent, yet the 0.53 per cent upper bound fails a 0.5 per cent gate. Would your AI feature pass?

Edition in preparation Read in Kai Books ↗ All books

No downloads, no subscriptions, nothing extra to buy: every numbered item in this book is printed in full and completed for a worked organisation, so you can adapt it to your own organisation straight from the page.

Premium Practitioner Edition. A complete working reference, not a quick read: practitioner chapters, a field toolkit of worked case studies, and a reference section with practice questions, checklists, an executive summary and an operating model you can adopt.

Twelve fabrications in 3,920 items look like 0.31 per cent, yet the 0.53 per cent upper bound fails a 0.5 per cent gate. Would your AI feature pass?

Most AI features fail for product reasons, not model reasons: nobody measured the baseline, tested the data, designed the fallback, set a threshold before the demo or modelled the cost of each successful answer. This book gives you an evidence-led method from first idea to retirement.

Twelve chapters follow the order real decisions are made, from opportunity sizing and discovery to feature design, evals, prototyping, model and cost choices, trust requirements, pricing, staged launch, outcome measurement and the operating model. Forty worked case files (CF-01 to CF-40), each a different feature at a different organisation, are tagged against fourteen patterns and indexed in full, and twenty-four completed toolkit artefacts follow one anchor feature from sizing to post-launch review.

Inside you will find:

• Forty worked AI feature case files, each tagged by pattern and indexed in one place

• Sizing, discovery and data readiness methods that kill weak ideas before engineering

• Golden sets, rubrics and Wilson-bound thresholds that make an eval number mean something

• Pricing corridors, power-user stress tests and unit economics for pay-per-use costs

• Ring rollouts, kill switches, outcome metrics and post-launch reviews with real verdicts

Part II is a field toolkit: two extended case studies, ten hands-on labs, fourteen completed templates, a nine-dimension maturity model with a thirty-metric pack and a ninety-day plan. Each element is designed to be lifted straight into your own programme and adapted to your organisation.

Part III adds a standards and regulatory crosswalk, six tabletop exercises, sixty practice questions, thirty failure patterns with recovery steps and a tooling and vendor evaluation guide. Consolidated checklists in the appendix bring every chapter together for quick reference.

Who it is for: AI product managers, heads of product and product leads shipping AI features, with the evaluation, engineering, finance and trust colleagues who decide alongside them.

Part of the Professional Validation Library by Professor Kai London: practical books that turn standards, technology and regulation into work you can evidence. The print editions include twelve framework figures, key takeaways for every chapter, a RACI matrix, a quick eight-row maturity scorecard mapped to the nine-dimension Capability Ladder, a 90-day plan, a list of abbreviations and a full index.

Professor Kai London