This site uses analytics cookies (Google Analytics, Microsoft Clarity) to understand how visitors use it — no advertising tracking. You can read the full Cookie Policy before deciding.
Reading List
The books that shaped how I think about AI strategy, business analysis, and product work — with honest notes on why each one earns a place on the list.
Teresa Torres
The definitive guide to building a continuous discovery practice — weekly customer touchpoints, opportunity solution trees, and assumption testing. Essential for product teams working on AI features.
Why I recommend it
“The opportunity solution tree is the best framework I know for mapping AI use cases against business outcomes.”
Marty Cagan
The product management canon. Cagan's model of empowered product teams is the baseline against which most AI product failures can be explained.
Why I recommend it
“Every AI initiative I've seen fail was a feature team problem, not a technology problem.”
Daniel Kahneman
The cognitive science foundation for understanding how users will actually interact with AI-driven products — biases, heuristics, and the limits of rational decision-making.
Why I recommend it
“You cannot design good AI experiences without understanding how people actually make decisions under uncertainty.”
Eric Ries
Build-measure-learn applied to AI projects changes everything — the biggest mistake AI teams make is skipping the validated learning loop entirely.
Why I recommend it
“Minimum viable AI experiments prevent teams from building six months of infrastructure before testing a hypothesis.”
Richard Rumelt
Rumelt's kernel of strategy — diagnosis, guiding policy, coherent actions — is the clearest mental model for separating real AI strategy from AI theatre.
Why I recommend it
“Most AI roadmaps I review are fluff, not strategy. This book teaches you the difference.”
Ralph Kimball & Margy Ross
The dimensional modelling bible. Understanding how data is structured for analytics is prerequisite knowledge for anyone specifying requirements for AI systems.
Why I recommend it
“AI requirements without data modelling knowledge produce systems that can never be trained correctly.”
Kai-Fu Lee
The competitive and geopolitical landscape of AI implementation. Useful context for understanding which AI use cases are defensible and which are commoditising fast.
Why I recommend it
“Understanding the global AI race shapes which capabilities to build in-house versus which to buy.”
Agrawal, Gans & Goldfarb
The most business-grounded framework for understanding what AI actually is — dramatically cheaper prediction — and how to identify where that changes a business model.
Why I recommend it
“The single best framework for a non-technical executive asking 'where should we use AI?'”
Jeff Patton
Story mapping is the most effective technique for aligning AI product scope with real user journeys. Patton's book is the definitive reference.
Why I recommend it
“Every AI product spec I write starts with a story map. It catches scope creep before it starts.”
Melissa Perri
Perri's diagnosis of output-focused organisations is the exact pathology that makes AI projects fail — teams shipping features that don't move metrics.
Why I recommend it
“AI teams are especially vulnerable to the build trap because the technology feels exciting regardless of outcome.”
Chris Voss
Negotiation techniques from an FBI hostage negotiator. Directly applicable to stakeholder elicitation, requirements conflict resolution, and getting senior buy-in for AI initiatives.
Why I recommend it
“The calibrated question technique changed how I run requirements workshops.”
John Doerr
OKRs as a measurement framework for AI initiatives — the book's case studies show how outcome-focused goal setting prevents the 'we built the AI but didn't move the metric' outcome.
Why I recommend it
“Every AI engagement should have OKRs defined before development starts.”
Rob Fitzpatrick
The shortest path to learning whether your customers actually have the problem you think you're solving. Essential pre-reading before any AI requirements phase.
Why I recommend it
“The number of AI projects that skip customer validation is astonishing. This book makes the cost visible.”
Charles Duhigg
Habit loops — cue, routine, reward — are the UX infrastructure for AI-powered behaviour change products. Understanding them at a neurological level makes requirements sharper.
Why I recommend it
“If your AI product is trying to change user behaviour, you need this framework.”
Cal Newport
The cognitive case for protecting time for high-quality analytical thinking. Directly relevant to how business analysts and strategists should structure their work.
Why I recommend it
“Requirements analysis requires the kind of focused thinking that most meeting-heavy teams structurally prevent.”
Get frameworks, not just book titles
Business-First AI™ Weekly applies the ideas in these books to real AI product and strategy challenges — every Thursday, free.
Subscribe to the newsletter →