Chapter 5: Product intent: Reverse-engineer the PRD
This introduction frames product-intent reverse engineering as using AI and source-code evidence to reconstruct a product brief that explains why a codebase exists.
After this chapter you will be able to
- Product perspective over implementation. Explain why deep fluency in implementation details can obscure the product purpose, and distinguish a technical description of a codebase from a description of what users use the product for.
- Code-backed product brief. Describe how source code can be used to reverse-engineer a product requirements document (PRD): implemented features are evidence of the team's product bets, often more candid than README or marketing copy.
- Three-prompt product analysis. State the chapter's three-part AI workflow for extracting product intent from code: produce an “It’s X, but Y” product pitch, map the competitive bet, and uncover details that marketing omitted.
- Curse of Knowledge. Explain how the Curse of Knowledge makes engineers see implementation mechanisms while users experience the relief of a problem being removed. Describe why deeper technical fluency can widen, rather than close, the gap between an engineer's view and a new user's view.
- User pain as a scene. Describe product intent by framing the pre-product pain as a concrete scene with one person, one moment of frustration, and the manual workaround they rely on. Distinguish this from an abstract problem summary, a feature list, or an architecture diagram.
- Engineer and user descriptions. Describe the two honest descriptions of a codebase: the engineer’s literal account of what mechanisms the code performs and the user’s account of the frustration the product eliminates. Explain why the technical description can be accurate yet inadequate as a product explanation.
- Universal product framing. Explain why every repository, including a library, CLI, or data tool, can be read as a product trying to win a specific customer away from a specific alternative.
- Delivery-model intent clues. Describe the four delivery models—web apps, libraries, CLIs, and data tools—and explain how the way a person interacts with each helps identify its customer and where to look for product intent.
- SaaS and developer-tool evidence. Distinguish where product intent is found in SaaS applications versus developer tools: reverse-engineer it from schema and UI in SaaS, but treat a developer tool's README pitch as a claim to verify against source code.
- Product one-liner. Describe a product in an “It’s X, but Y” sentence that anchors it to a familiar product and states its one concrete differentiating bet. Explain why this format avoids a tech-stack summary and helps a stranger picture what to build.
- Ideal customer profile. Describe how to infer the ideal customer profile (ICP) from who pays for and uses the product, and identify the consumer, prosumer, business-owner, enterprise, developer, and two-sided-marketplace profiles discussed in this section. Explain that the ICP reveals where the engineering team concentrates its effort.
- Customer signals in code. Trace the repository evidence that supports an ICP hypothesis, such as consumer polish and social login, prosumer power features, enterprise identity controls, developer APIs and SDKs, or marketplace flows for both sides. Explain how these signals reveal which experience receives the most engineering attention.
- Positioning trade-off analysis. Explain positioning as the specific trade a product makes against alternatives, including what it deliberately sacrifices and what it gains. Describe why a comparison should reveal meaningful splits between competitors rather than list generic shared features.
- Counter-positioning mechanism. Describe counter-positioning as a strategy in which a product does something an incumbent cannot copy without damaging its existing business model. Explain Cal.com's white-labeling trade: it gives up viral exposure from branded booking links in exchange for deep customer embedding and a position Calendly cannot copy without weakening its referral loop.
- Positioning comparison prompt. Describe the prompt structure for asking an LLM to analyze positioning: compare the product with three or four competitors across four meaningful dimensions, define the dimensions in plain English, use a table, and conclude with the sacrifice, gain, and whether the position is defensible. Explain that the analysis should lead with user or product meaning rather than code evidence.
- Moat and wedge tests. State the four moats—switching costs, network effects, brand, and counter-positioning—and explain how the moat should guide engineering investment. Describe the wedge as the narrow beachhead customer for whom the product is irresistible, using Cal.com's embedded scheduling need for startups such as healthcare or recruiting products.
- Hidden product bets. Explain how a capability that exists in the code but is absent from marketing can reveal a hypothesis about the product’s next move. Describe how Cal.com’s Agent and CalAiPhoneNumber models suggest proactive AI-driven scheduling rather than a passive booking form.
- Deliberate strategic gaps. Describe how a conspicuously absent feature, especially one rivals provide, can signal strategic focus and the customer a team is willing to disappoint. Explain that an omission becomes persuasive when it survives user requests, competitor pressure, and alternative ways to solve the problem.
- Strategic signal investigation. Apply the section’s investigation prompt by producing two tables: surprisingly present capabilities and deliberately absent competitor capabilities. State each finding as a direct noun-verb fact that foregrounds the user or product meaning and includes a visible path, table name, or configuration flag as evidence.
What it builds on
- The reframe: From reading code to architectural thinking (chapter 1) · The reframe from merely reading code to architectural thinking.
- What the LLM sees determines everything (chapter 3) · The importance of what repository context the LLM sees.
- Skill file: Make the agent document what it learned (chapter 4) · Using AI to document findings from codebase exploration.
- A concrete prompt beats a vague one, every time (chapter 2) · The value of concrete prompts over vague requests.
- Swap the instructions, change the output (chapter 3) · Changing instructions to analyze the same codebase through different lenses.
- Build and run: Ground truth over guesses (chapter 4) · Using builds and runs as ground truth instead of relying on guesses.
Section by section
Introduction
This introduction frames product-intent reverse engineering as using AI and source-code evidence to reconstruct a product brief that explains why a codebase exists.
- Product perspective over implementation: Explain why deep fluency in implementation details can obscure the product purpose, and distinguish a technical description of a codebase from a description of what users use the product for.
Why it matters: Understanding the product purpose keeps architectural and code-level details connected to the reason the software exists. - Code-backed product brief: Describe how source code can be used to reverse-engineer a product requirements document (PRD): implemented features are evidence of the team's product bets, often more candid than README or marketing copy.
Why it matters: Code lets you ground a product story in what the team actually built rather than in a polished public description. - Three-prompt product analysis: State the chapter's three-part AI workflow for extracting product intent from code: produce an “It’s X, but Y” product pitch, map the competitive bet, and uncover details that marketing omitted.
Why it matters: The three prompts turn a large unfamiliar repository into a concise, evidence-based product story.
Key terms: Curse of Knowledge, product, implementation details, source code, product requirements document (PRD), product brief, features, product bets, three prompts, It's X, but Y
5.1 Why experts misread the products they build
This section explains why implementation knowledge obscures product intent and introduces user-pain scenes as a way to recover the real problem a codebase solves.
- Curse of Knowledge: Explain how the Curse of Knowledge makes engineers see implementation mechanisms while users experience the relief of a problem being removed. Describe why deeper technical fluency can widen, rather than close, the gap between an engineer's view and a new user's view.
Why it matters: If you only see tables, workers, and routes, you can miss the reason anyone chooses the product. - Product work beyond mechanisms: Distinguish familiar technical building blocks from the product work that makes an experience trustworthy, effortless, and valuable to non-technical users. Explain why dismissing a successful product as “just a wrapper” overlooks the work users actually pay for.
Why it matters: Recognizing the difference keeps you from mistaking an easily named mechanism for the whole product. - User pain as a scene: Describe product intent by framing the pre-product pain as a concrete scene with one person, one moment of frustration, and the manual workaround they rely on. Distinguish this from an abstract problem summary, a feature list, or an architecture diagram.
Why it matters: A concrete before-state reveals the problem the product exists to remove and makes its real competitor visible. - Engineer and user descriptions: Describe the two honest descriptions of a codebase: the engineer’s literal account of what mechanisms the code performs and the user’s account of the frustration the product eliminates. Explain why the technical description can be accurate yet inadequate as a product explanation.
Why it matters: You need both readings to avoid confusing a correct account of code with an explanation of product value.
Key terms: Curse of Knowledge, implementation details, user's view, wrapper, building blocks, trustworthy, effortless, pain, scene, manual workaround
5.2 Every codebase builds for someone, even a library
This section frames every repository as a product for a customer, whose delivery model and artifacts reveal the pain, alternative, and intent it is trying to win against.
- Universal product framing: Explain why every repository, including a library, CLI, or data tool, can be read as a product trying to win a specific customer away from a specific alternative.
Why it matters: This framing turns unfamiliar code from a collection of files into evidence of a product bet. - Delivery-model intent clues: Describe the four delivery models—web apps, libraries, CLIs, and data tools—and explain how the way a person interacts with each helps identify its customer and where to look for product intent.
Why it matters: Classifying a repository quickly narrows the people, workflow, and evidence that matter when you begin reading it. - SaaS and developer-tool evidence: Distinguish where product intent is found in SaaS applications versus developer tools: reverse-engineer it from schema and UI in SaaS, but treat a developer tool's README pitch as a claim to verify against source code.
Why it matters: Using the right evidence prevents you from treating marketing claims as proof or searching the wrong artifacts for the product pitch. - Intent-extraction sequence: Trace the chapter's three-step intent analysis: write a one-sentence product definition, map the product against rivals, and identify constraints that the code reveals but marketing hides.
Why it matters: The sequence provides a structured way to build a product mental model instead of making a vague guess from isolated files.
Key terms: customer, pain, incumbent, product bet, alternative, web app, library, command-line interface (CLI), data tool, delivery model
5.3 Ask for the product in one "It's X, but Y" sentence
This section shows how to infer a codebase’s product positioning and ideal customer by producing a concrete “It’s X, but Y” description and checking what the repository prioritizes.
- Product one-liner: Describe a product in an “It’s X, but Y” sentence that anchors it to a familiar product and states its one concrete differentiating bet. Explain why this format avoids a tech-stack summary and helps a stranger picture what to build.
Why it matters: A concrete one-liner turns unfamiliar code into a product you can reason about rather than a list of technologies. - Reproduce-it test: Explain the reproduce-it test for a product description: a stranger should be able to use the sentence to start rebuilding the core product. Distinguish specific, buildable descriptions from vague promotional language.
Why it matters: The reproduce-it test lets you reject impressive-sounding descriptions that reveal no actionable product intent. - Ideal customer profile: Describe how to infer the ideal customer profile (ICP) from who pays for and uses the product, and identify the consumer, prosumer, business-owner, enterprise, developer, and two-sided-marketplace profiles discussed in this section. Explain that the ICP reveals where the engineering team concentrates its effort.
Why it matters: Knowing who the product is trying to win helps you interpret why particular subsystems are polished, complex, or absent. - Customer signals in code: Trace the repository evidence that supports an ICP hypothesis, such as consumer polish and social login, prosumer power features, enterprise identity controls, developer APIs and SDKs, or marketplace flows for both sides. Explain how these signals reveal which experience receives the most engineering attention.
Why it matters: Repository signals give you evidence for product intent instead of relying on README claims or your first impression.
Key terms: It’s X, but Y, familiar anchor, product bet, tech-stack summary, reproduce-it test, relentless specificity, promotional language, ideal customer profile, ICP, consumer
5.4 Competitive positioning: The one bet rivals can't copy
This section shows how to use competitor comparisons to uncover a product's defensible strategic trade-off, the moat behind it, and the narrow customer segment it wins first.
- Positioning trade-off analysis: Explain positioning as the specific trade a product makes against alternatives, including what it deliberately sacrifices and what it gains. Describe why a comparison should reveal meaningful splits between competitors rather than list generic shared features.
Why it matters: A product's strategic intent is often clearer in what it refuses to do than in its feature list. - Counter-positioning mechanism: Describe counter-positioning as a strategy in which a product does something an incumbent cannot copy without damaging its existing business model. Explain Cal.com's white-labeling trade: it gives up viral exposure from branded booking links in exchange for deep customer embedding and a position Calendly cannot copy without weakening its referral loop.
Why it matters: Counter-positioning helps you identify differences that are structurally defensible rather than temporarily novel. - Positioning comparison prompt: Describe the prompt structure for asking an LLM to analyze positioning: compare the product with three or four competitors across four meaningful dimensions, define the dimensions in plain English, use a table, and conclude with the sacrifice, gain, and whether the position is defensible. Explain that the analysis should lead with user or product meaning rather than code evidence.
Why it matters: A focused prompt steers the LLM away from shallow feature inventories and toward the strategic divide in the codebase. - License business-model link: Explain how an open-source license can make a positioning bet financially viable. Describe how Cal.com's AGPLv3 copyleft requirement pushes network-service users who want to keep modified code private toward a paid license, helping fund its white-label strategy.
Why it matters: The LICENSE file can reveal the economic mechanism supporting a product strategy before pricing pages or marketing copy do. - Moat and wedge tests: State the four moats—switching costs, network effects, brand, and counter-positioning—and explain how the moat should guide engineering investment. Describe the wedge as the narrow beachhead customer for whom the product is irresistible, using Cal.com's embedded scheduling need for startups such as healthcare or recruiting products.
Why it matters: Knowing the moat and wedge helps predict where a company will invest and which users its product choices are designed to win.
Key terms: positioning, trade-off, feature checklist, competitive dimensions, counter-positioning, incumbent, white-labeling, viral growth loop, Powered by Calendly footer, referral loop
5.5 Read what the code admits that the marketing won't
This section shows how to infer a product’s hidden strategy from unadvertised capabilities, deliberate omissions, and half-built commitments visible in its codebase.
- Hidden product bets: Explain how a capability that exists in the code but is absent from marketing can reveal a hypothesis about the product’s next move. Describe how Cal.com’s Agent and CalAiPhoneNumber models suggest proactive AI-driven scheduling rather than a passive booking form.
Why it matters: Hidden capabilities can reveal where a company is investing before its public messaging catches up. - Deliberate strategic gaps: Describe how a conspicuously absent feature, especially one rivals provide, can signal strategic focus and the customer a team is willing to disappoint. Explain that an omission becomes persuasive when it survives user requests, competitor pressure, and alternative ways to solve the problem.
Why it matters: What a team repeatedly refuses to build can expose its strategy more clearly than its feature list does. - Strategic signal investigation: Apply the section’s investigation prompt by producing two tables: surprisingly present capabilities and deliberately absent competitor capabilities. State each finding as a direct noun-verb fact that foregrounds the user or product meaning and includes a visible path, table name, or configuration flag as evidence.
Why it matters: A structured evidence table turns vague impressions from a repository into testable claims about product strategy. - Verification of structural evidence: Describe how to treat hidden capabilities and missing features as hypotheses rather than conclusions. State that hidden bets should be checked for repeated investment, wiring, and recent activity, while gaps should be corroborated through the schema, integrations, and configuration.
Why it matters: Verification distinguishes a real strategic commitment from backlog noise, accidental absence, or dead code. - Half-built roadmap signals: Explain how partially built structures can reveal roadmap commitments before a feature ships. Identify empty configuration slots, stub database tables, and lightly consumed worker queues as examples of “poured concrete,” including Cal.com’s DAILY_API_KEY slot, half-built tasker queue, and expanding Team table.
Why it matters: Resources already embedded in the codebase often predict future product work better than a public coming-soon message.
Key terms: hidden bet, unadvertised capability, Agent, CalAiPhoneNumber, deliberate gap, strategic focus, target customer, trade-off, what's surprisingly present, what's deliberately absent
5.6 Summary
This summary recaps how to infer a product’s user problem, positioning, and hidden strategic bets from its code rather than from feature lists or marketing.
Where this is used later
- Git log: The record nothing else keeps (chapter 6) · It establishes the broader practice of recovering product intent from repository evidence.
- Read the graveyard of killed features (chapter 6) · Expressing product intent as a concise strategic statement.
- Map the whole API surface into a feature menu (chapter 8) · A feature menu of the API can reveal developer-oriented product capabilities and customer priorities.
- Ask the agent to inventory every part from the config (chapter 9) · Inventorying system parts from configuration can provide further evidence of the capabilities the product invests in.
- Ask the LLM to draw the schema and tour it (chapter 7) · Schema tours help surface models and tables that reveal hidden product capabilities.
- Summary (chapter 6) · the practice of reading what the code reveals beyond what marketing claims
- Three design questions to ask once you have the menu (chapter 8) · An API feature menu supports reasoning about which product capabilities are present or intentionally missing.
- Aggregate commits by directory and month into eras (chapter 6) · The same evidence-oriented approach can aggregate code changes into product eras.
Concepts explained in this chapter
- market positioningPositioning is the strategic way a product is framed against alternatives by emphasizing a distinctive value and trade-off for a specific market.
- counter-positioningCounter-positioning is a competitive strategy in which a company adopts an approach that an incumbent cannot readily copy without damaging its existing business.
- reproduce-it testThe reproduce-it test is a product-description criterion asking whether a reader is sufficiently informed to build a copy of the described product.
- network effectA network effect occurs when each additional user increases the value of a product or service for other users.
- switching costA switching cost is the effort, expense, or disruption a customer incurs when replacing one product or provider with another.
- wedgeA wedge is a narrow initial customer segment or use case that a product aims to win decisively before expanding.
- white-label customizationWhite-labeling is the practice of removing a provider's branding from a product so another organization can present it as its own.
- X-but-Y product descriptionAn X-but-Y product description defines a product by anchoring it to a familiar reference product and stating one concrete differentiating choice.
- ideal customer profileAn ideal customer profile is a description of the customer segment a product is designed to attract and serve.
- product betA product bet is the deliberate strategic choice that differentiates a product through its target customer, exclusions, or method of delivering value.
- Curse of KnowledgeThe Curse of Knowledge is a cognitive bias in which knowing how something works makes it difficult to imagine the perspective of someone who does not know it.
- deliberate gapA deliberate gap is a capability that a product intentionally does not provide despite its availability in competing products.
- hidden featureA hidden feature is a built product capability that is not publicly advertised.
- roadmap signalA roadmap signal is partially implemented code or configuration that indicates an unannounced or unfinished product direction.
- strategic signalA strategic signal is an observable product or codebase characteristic that indicates a team's priorities and future direction.
- command-line interfaceA CLI is a command-line interface program that accepts user commands and produces textual output.
- libraryA software library is reusable code packaged for incorporation into another program.
- product intentProduct intent is the intended customer, problem, competitive alternative, and design position that explain why a software product exists.
- artificial intelligenceArtificial intelligence is the field of building computational systems that perform tasks requiring capabilities such as reasoning, language understanding, or pattern recognition.
- promptA prompt is an instruction or input supplied to an AI model to elicit a particular response.
- repositoryA repository is a structured collection of a software project's files and directories managed as a unit.
These guides are generated from the text of Manning books by Manning's knowledge graph and link to the sections they draw on. The full books are available on liveBook and at manning.com.
