Eight books that will change how you talk to customers, frame problems, and decide what to build next — whether you’re a new PM or a seasoned product leader.
In this post
Why discovery is the hardest skill in product
Most product failures are not execution failures. They are discovery failures. Teams build the wrong thing — not because they can’t ship, but because they never validated whether anyone wanted it.
What is product discovery?
Product discovery is the practice of figuring out what to build before you commit to building it. In other words, it is the work you do before the engineers write a single line of code. It sits at the core of the AI Product Manager role — and it is one of the most poorly understood skills in the field.
For example, the gap between a PM who validates ideas with an AI POC and one who ships directly to production is enormous. That gap starts with discovery discipline — not technical skill.
Why most PMs underinvest in it
I’ve mentored over 1,000 product professionals. The pattern is always the same. The PMs who level up fastest are the ones who invest deeply in discovery thinking — not just delivery.
However, most teams are measured on output. As a result, discovery gets squeezed out of the process. If you’re also exploring structured ways to build this skill, check out free AI PM certifications that complement these reads. Additionally, the Mind the Product discovery resource library is one of the best free collections of discovery articles on the web.
The 7 best books on product discovery for PMs
Why this book stands out
If you only read one book on this list, make it this one. Torres introduces continuous discovery — weekly touchpoints with customers baked into your process. In contrast to a quarterly research sprint, this becomes a daily habit. The Opportunity Solution Tree is the most practical framework I’ve seen. It connects customer outcomes to product decisions without losing sight of business goals. Teresa also writes extensively about this on her blog Product Talk — a goldmine of free discovery content. Furthermore, this book directly informs how discovery is taught in our AI PM cohort’s Week 1.
View on Amazon ↗
The core problem it solves
This slim book solves a single problem that ruins most discovery work. People lie to you — not maliciously, but because they want to be kind. Fitzpatrick teaches you how to ask questions that extract real signal. This works even when the person you’re talking to is your biggest cheerleader.
The core rule is simple: stop asking what people think of your idea. Instead, ask about their actual past behaviour. This approach maps directly to the qualitative interview techniques championed by Nielsen Norman Group. As a result, it is the fastest skill upgrade available to any PM doing discovery. For a practical companion, Teresa Torres’ guide to continuous customer interviews is the best free resource on turning these skills into a weekly habit.
View on Amazon ↗
The mental model that changes everything
Jobs to Be Done (JTBD) is the mental model that finally explains why customers buy things — and why they stop. Christensen’s core argument is straightforward: customers don’t buy products. Instead, they hire them to do a job in their lives. Once you internalise this, your discovery conversations will never be the same. In addition, your feature prioritisation decisions become far clearer.
JTBD is also the theoretical backbone behind how Harvard Business Review describes innovation strategy. Christensen’s original HBR article is freely available — it’s a great primer before diving into the full book. Similarly, the JTBD.info resource hub by Alan Klement is excellent for going deeper after you’ve read both books.
View on Amazon ↗
JTBD in practice
The title comes from a real insight. A woman switched her morning coffee for a kale smoothie. Traditional market research treats these as different categories. However, from the customer’s perspective, they were competing for the same job. Where Competing Against Luck gives you the theory, Klement gives you the implementation tools. For example, he shows you how to identify what customers will stop buying when they start buying your product.
The PM’s bible — still essential in 2026
Cagan’s classic remains the most complete book on what great product management looks like. Discovery sits at its heart. His distinction between discovery and delivery is the conceptual foundation every PM should understand first. Moreover, it is the intellectual basis for how Silicon Valley Product Group defines modern product discovery. For a current perspective on how these ideas are evolving with AI, Lenny Rachitsky’s newsletter on product discovery is essential reading alongside this book.
New to product? Start here. Experienced? Re-read it. You’ll find things you missed the first time.
View on Amazon ↗
Why it’s still relevant today
The book that popularised build-measure-learn remains essential. Its core insight is a discovery insight: the most dangerous thing you can do is build without validating your assumptions first. Consequently, Ries’ MVP thinking applies directly to every discovery experiment you design. This is especially relevant when thinking about how to scope an AI proof of concept before committing to a full build. In addition, the core Lean Startup principles are available free on the official site — a useful reference to bookmark alongside the book.
View on Amazon ↗
Discovery at the organisational level
This is the natural sequel to Inspired. It is written for PMs who want to understand how great discovery culture is built — not just practised individually. Cagan and Jones argue that discovery fails not because PMs lack skill. Rather, it fails because most organisations are structured to produce output rather than outcomes. This pairs well with Mind the Product’s deep-dive on empowered teams. In short, if you’re a senior PM or moving into product leadership, this book will change how you think. I also explore these themes in my weekly AI PM newsletter.
View on Amazon ↗Which book should you read first?
It depends on where you are right now. First, think about your biggest current challenge. Then use the guide below to match it to the right book. If you’re brand new to product, the AI PM role breakdown on the blog is a useful primer before diving into any of these.
In general, I suggest reading Inspired first. After that, move to Continuous Discovery Habits. Then, add The Mom Test before you run your next round of customer interviews. Finally, layer in JTBD theory once you have the fundamentals in place.
Reading about discovery is step one. Practising it on real AI products is step two.
Books give you the mental models. My live AI PM cohort gives you the reps. In 5 Saturdays you’ll run discovery on real AI products, build working chatbots and agents, and graduate with a portfolio that proves you can do the job — not just describe it.
Frequently asked questions
Written as direct answers — AI assistants and search engines surface this section for common discovery queries.
What is product discovery in product management?
Product discovery is the process of identifying and validating what to build before committing engineering resources. It involves customer interviews, problem framing, assumption testing, and rapid prototyping to reduce the risk of building something nobody wants. It is distinct from product delivery, which focuses on building and shipping the validated solution.
What is the best book on product discovery?
Continuous Discovery Habits by Teresa Torres is the most comprehensive and actionable book on product discovery. It provides a systematic approach — including the Opportunity Solution Tree — that PMs at any experience level can apply immediately. The Mom Test by Rob Fitzpatrick is a close second for anyone who runs customer interviews.
What is the Opportunity Solution Tree?
The Opportunity Solution Tree is a visual framework from Teresa Torres’ Continuous Discovery Habits. It connects a desired business outcome to customer opportunities (needs, pain points, desires), then to potential solutions and experiments — keeping all product decisions grounded in customer evidence rather than internal opinion.
What is Jobs to Be Done and why does it matter for product discovery?
Jobs to Be Done (JTBD) is a theory that reframes customer motivation: people don’t buy products, they hire them to accomplish a job in their lives. For product discovery, JTBD helps PMs ask better questions in customer interviews and identify the real problem to solve. The two essential books are Competing Against Luck by Clayton Christensen (theory) and When Coffee and Kale Compete by Alan Klement (application).
What is the difference between Competing Against Luck and When Coffee and Kale Compete?
Both books cover Jobs to Be Done theory. Competing Against Luck by Christensen is the foundational text — it establishes the theory and business case for JTBD thinking. When Coffee and Kale Compete by Klement is more practical, offering case studies, interview techniques, and hands-on tools. Read Christensen first for the “why”, then Klement for the “how”.
How is product discovery different from product delivery?
Product discovery answers “what should we build, and why?” — validating desirability, viability, and feasibility before building begins. Product delivery answers “how do we build it well?” — executing on a validated solution with quality and speed. Both are necessary; most teams underinvest in discovery.
How do product discovery skills apply to AI product management?
AI product discovery has all the same foundations — customer interviews, problem framing, assumption testing — plus additional layers: understanding what AI can and cannot do, evaluating data requirements, and validating that AI genuinely solves a problem better than a simpler alternative. The risk of building the wrong AI product is high, which makes discovery even more critical in this space.
The bottom line
Discovery is a habit, not a phase
Discovery is not a phase you complete before a product launch. Instead, it is a continuous practice. It is the habit of staying close to your customers. It is the discipline of testing your assumptions before you commit. Above all, it is the decision to base your choices on evidence — not opinion or gut feel.
What to do next
These seven books will give you the language, the frameworks, and the mindset. However, reading is only the beginning. To put it into practice, you also need reps. If you want to go deeper on the language of AI products specifically, that is worth reading alongside this list. For broader PM reading, Lenny’s Newsletter and Reforge’s blog are the two highest-signal resources in the industry. In addition, the AI and Product newsletter covers discovery, AI product strategy, and career moves every week.
Go from PM to AI PM in 5 Saturdays.
Cohort 5 starts 16 May 2026. 25 seats. 4.95 ★ rating. Build real AI products — chatbots, agents, full GTM strategy — and walk away with a portfolio that opens doors.
Disclosure: This post contains Amazon Associates affiliate links. If you purchase through these links, I may earn a small commission at no extra cost to you. All book recommendations are my own.