Product Sense
You are in a meeting. The data is ambiguous. User research points in one direction, the market in another, and your CEO in a third. There is no framework that will give you the answer. This is the moment where product sense — the ability to make good product decisions without complete information — separates PMs who build great products from PMs who build adequate ones. The question every PM eventually confronts is: can you develop this ability, or are some people simply born with better product instincts?
The Core Idea
Product sense is the ability to understand what makes a product good — to look at a product, a feature, a user flow, or a market opportunity and intuitively assess whether it will work, why it will or will not resonate with users, and how it could be better. It is the judgment layer that sits above data, frameworks, and processes.
Mihika Kapoor, product manager at Figma and formerly at Meta, draws on how Meta distilled the PM role: “Meta basically distilled the product role into two core capabilities. One was product sense and one was execution. And when you think about product sense, it’s like, okay, what is product sense? It’s like a really abstract term. And at the end of the day, I think product sense is just having good intuition.”
This definition is important because it frames product sense as a skill — pattern recognition built through experience — not as an innate gift. Patterns can be learned. The PM who has seen 50 onboarding flows can spot a problematic one faster than the PM who has seen 5. The PM who has watched three companies try and fail to move upmarket recognizes the warning signs when their company starts the same journey.
But Shreyas Doshi, speaking at the Lenny and Friends Summit, adds an important caveat about taste that goes beyond product design: “We talk about taste, we all talk about taste, and it’s about the beautiful pixels, and the perfect product. But I think there is something that we as product leaders needed to recognize about taste as just a factor in pretty much everything we do, which is like, do we have good taste around the beliefs we choose to create within ourselves as product leaders?” Taste, in his framing, is not just about UI polish — it is about whether you have good judgment about what to believe, who to learn from, and how to evaluate the ideas that come your way.
The Components of Product Sense
Product sense is not a single ability. It is a compound of several distinct capabilities, each of which can be developed independently.
The Four Components
| Component | What It Is | How You See It |
|---|---|---|
| User empathy | The ability to understand how users think, feel, and behave — especially users who are different from you | PM can predict user confusion before usability testing reveals it. Designs for real behavior, not ideal behavior. |
| Domain knowledge | Deep understanding of the market, competitors, technology, and business model | PM knows which competitive moves matter and which are noise. Understands why previous attempts at a solution failed. |
| Creative ability | The ability to generate novel solutions to product problems | PM proposes approaches that others had not considered. Combines ideas from adjacent domains. |
| Technical understanding | Enough knowledge of the technology to understand feasibility, cost, and architectural implications | PM does not propose solutions that are technically naive. Understands trade-offs between speed and scalability. |
Kapoor describes how she builds this compound ability in practice: “How do you build up intuition? I think that it’s just by having this insatiable curiosity and talking to users at every chance you get. Once you start having enough conversations, over time you build this almost repository or library of conversations that you can draw from as you’re making product decisions.” The four components develop not through theory but through accumulated exposure — and the PMs who synthesize that exposure most effectively are the ones who develop the strongest product sense.
How the Components Interact
Strong product sense requires all four components working together. A PM with strong user empathy but weak technical understanding will propose solutions that users love but that are prohibitively expensive to build. A PM with strong domain knowledge but weak creative ability will correctly diagnose problems but only propose conventional solutions.
| Missing Component | Resulting Failure Pattern |
|---|---|
| Weak user empathy | Builds features that make logical sense but that users find confusing, unnecessary, or misaligned with their actual workflow |
| Weak domain knowledge | Builds solutions that have already been tried and failed, or misreads competitive dynamics |
| Weak creative ability | Defaults to copying competitors or implementing the most obvious solution when a better approach exists |
| Weak technical understanding | Proposes solutions that are technically infeasible, or misses opportunities that the technology enables |
Can You Develop Product Sense or Is It Innate?
This question comes up repeatedly across episodes, and the consensus among guests is nuanced: most of product sense is developable, but the starting points vary.
Judd Antin, former head of research at Airbnb and Meta, frames product sense through the lens of how intuition actually works: “Great leaders develop intuition. It’s the pattern matching part of experience, where you develop heuristics which allow you to make good judgments even if you can’t quite explain where that judgment came from. That’s what the gut is. But it’s also where bias comes from, where all the cognitive biases live.” The implication: product sense develops through experience, but it requires calibration. Raw intuition without falsification is just confident guessing.
Bob Baxley, design leader at Apple, Pinterest, and ThoughtSpot, describes how to accelerate this development through deliberate observation: “Go watch people using software in the wild. Comedians can go to a comedy club, they can start to develop an intuition about why people laugh. None of us have an obvious way to go watch people use software, so we don’t really understand how humans process what’s happening on the screen. And you have to just find ways to do that.” The PMs who actively study how people interact with products — not just their own, but any software — build pattern recognition faster than those who rely solely on dashboards.
The practical answer: treat product sense as developable. Even if some component (taste, creative ability) has an innate ceiling, the ceiling is high enough that deliberate development produces significant returns for virtually every PM.
The Development Timeline
| PM Level | Typical Product Sense | What Is Developing |
|---|---|---|
| Junior PM (0-2 years) | Relies heavily on data and frameworks. Makes decisions that are safe but rarely surprising. | Building domain knowledge and user empathy through direct exposure |
| Mid-level PM (2-5 years) | Developing intuition in their domain. Can identify problems faster than solutions. | Pattern recognition forming from accumulated experience |
| Senior PM (5-10 years) | Strong intuition. Can predict user reactions, competitive moves, and technical implications. | Refining taste. Developing cross-domain pattern recognition. |
| Product leader (10+ years) | Sets product direction based on conviction. Makes bets that seem risky to others but are informed by deep pattern recognition. | Teaching and scaling product sense across the organization |
Evaluating Product Sense in PM Interviews
Product sense is the most sought-after and hardest-to-evaluate quality in PM hiring. Most PM interview processes include a “product sense” round, but the quality of evaluation varies enormously.
How to Evaluate Product Sense
Evaluating each component separately rather than relying on a single “product sense question” produces better signal:
| Component | Interview Question Type | What Good Looks Like |
|---|---|---|
| User empathy | ”Who are the users of [product]? Walk me through their experience.” | Candidate identifies distinct user segments, describes their motivations and constraints, and notes where their needs diverge |
| Domain knowledge | ”What is [product] doing well? What are they missing?” | Candidate demonstrates awareness of competitive landscape, market dynamics, and strategic context — not just surface-level feature comparison |
| Creative ability | ”How would you improve [product]?” | Candidate generates multiple ideas at different levels of investment, makes explicit trade-offs, and proposes at least one non-obvious approach |
| Technical understanding | ”How would you build this? What are the main technical considerations?” | Candidate identifies key technical trade-offs (build vs. buy, speed vs. scalability) without going so deep into implementation that they lose the product perspective |
Product Critique Exercise
A common interview format: give the candidate a product and ask them to evaluate it. The quality of the critique reveals product sense more reliably than hypothetical questions.
What weak product sense looks like in a critique:
- Lists features without analyzing why they exist
- Proposes improvements without identifying who they serve
- Focuses exclusively on UI details without addressing strategic questions
- Copies competitors without understanding why the competitor made that choice
- Cannot articulate trade-offs
What strong product sense looks like:
- Identifies the product’s core value proposition and evaluates how well the product delivers on it
- Distinguishes between different user segments and their distinct needs
- Proposes improvements connected to specific user outcomes, not just personal preferences
- Explains why the product team might have made certain choices, even when disagreeing
- Articulates trade-offs explicitly: “This would improve X but at the cost of Y”
Dylan Field, CEO of Figma, illustrates how strong product sense manifests at the leadership level through the story of launching FigJam. He describes how the decision to differentiate through fun — which the team initially resisted — came from intuition built through years of observing how people collaborate: “This is a place where I can definitely describe it all looking backwards, but if I’m going to be totally honest, at that time it was more intuition. I think I had an intuition that the value was moving up the stack.” The best signal in a product critique, whether in an interview or a product review, is not whether the ideas are good but whether the reasoning behind them is sound.
Pattern Recognition Across Companies
Product sense, at its core, is pattern recognition. PMs with strong product sense have internalized patterns from many products and can recognize when a current situation matches a pattern they have seen before.
Common Patterns That Strong PMs Recognize
| Pattern | What It Means | Example |
|---|---|---|
| The complexity trap | Adding features to serve more use cases while making the core experience worse for everyone | Evernote adding presentations, chat, and work chat while the core note-taking experience degraded |
| The empty room problem | Building a collaborative product but launching it to individual users who have no one to collaborate with | Slack is useless with one person. The go-to-market strategy must account for team adoption, not individual sign-ups |
| The power user trap | Optimizing for power users while making the product increasingly intimidating for new users | Jira’s interface complexity, which serves power users but terrifies new teams |
| The metrics mirage | A metric improving while the underlying user experience gets worse | Increasing time-in-app through confusing navigation rather than engaging content |
| The platform premature | Building a platform before building a single compelling use case | Building APIs and extensibility before anyone wants to use the core product |
| The local maximum | Small optimizations that improve the current approach but prevent discovering a fundamentally better approach | Optimizing a checkout flow when the real opportunity is eliminating checkout entirely (Amazon 1-click) |
Doshi warns that pattern recognition has a shadow side — being drawn to patterns for the wrong reasons: “We get overly excited about cool metaphors. We get very impressed with alliterations. Fail fast — what if that thing were called fail quickly? Do you think you would be as attracted to that idea?” True product sense means evaluating ideas on their merits, not their social proof, authority bias, or catchy framing. A PM who has seen the complexity trap play out at two previous companies will recognize it instantly at their current company. That recognition — which feels like intuition to others — is actually just well-organized memory, but only if it has been calibrated against results rather than built on borrowed conviction.
How to Practice and Improve Product Sense Deliberately
Product sense improves with exposure and reflection. Exposure alone is not enough — many PMs use hundreds of products without developing strong product sense. The difference is whether you analyze what you experience.
Daily Practices
| Practice | Time Investment | What It Develops |
|---|---|---|
| Product teardowns | 15-30 min, 2-3x/week | Analyze a product you use: What is the core value? Who is it for? What is the growth model? What would you change? |
| User observation | 30 min/week | Watch a user (colleague, friend, family member) use a product. Note what confuses them, delights them, or causes them to give up. |
| Decision journaling | 10 min/day | Before each product decision, write your prediction. After the outcome, compare. Where were you right? Where were you wrong? Why? |
| Competitive analysis | 1 hour/week | Study a competitor’s recent changes. Why did they make that decision? What problem does it solve? What trade-off did they accept? |
| Cross-domain study | 30 min/week | Study products outside your domain. How does Duolingo’s onboarding apply to your B2B tool? What can a marketplace PM learn from a gaming PM? |
Structured Development
For PMs who want to make step-function improvements, a more structured approach helps:
Quarter 1: Audit your components. Rate yourself on user empathy, domain knowledge, creative ability, and technical understanding. Ask your manager, designer, and engineer to rate you as well. Identify your weakest component.
Quarter 2: Deliberate practice on the weak component. If user empathy is weak, double your user interviews and practice predicting user behavior before testing. If domain knowledge is weak, read everything your competitors publish and interview industry experts. If creative ability is weak, force yourself to generate five solutions to every problem before evaluating any of them.
Quarter 3: Integrate and calibrate. Apply the improved component to real product decisions. Track your predictions against outcomes. Adjust.
Quarter 4: Teach it. Explain your reasoning to junior PMs. Teaching forces you to articulate the patterns you have internalized, which strengthens them.
Product Sense vs. Data Literacy
Both are essential for modern product management. But they are different capabilities, and confusion between them leads to two distinct failure modes.
| Dimension | Product Sense | Data Literacy |
|---|---|---|
| What it answers | ”Is this a good idea?" | "Did this idea work?” |
| When it matters most | Generating and evaluating options | Measuring and validating outcomes |
| Risk when missing | Build the wrong things with great execution | Build the right things but cannot prove or improve them |
| How it feels | ”I think users will struggle with this flow" | "Step 3 has a 40% drop-off rate among mobile users” |
| Development path | Exposure, reflection, pattern recognition | Statistics, analytics tooling, experimental design |
The Two Failure Modes
Product sense without data literacy: The PM with great instincts who cannot validate them. They “know” what to build, ship it, and claim success without measurement. When challenged, they appeal to conviction rather than evidence. They are right more often than average — but they cannot prove it, and when they are wrong, they do not learn from it.
Data literacy without product sense: The PM who can analyze any dataset but cannot generate an original product idea. They optimize existing flows effectively but cannot envision new ones. They are excellent at incremental improvement and terrible at step-function innovation. They default to “what does the data say?” in situations where the data is silent, ambiguous, or misleading.
Antin frames the tension between intuition and data as a false binary: “There is an important place for intuition in product development, of course. The best designers, researchers, product people develop strong intuition for the product. But you got to understand, intuition is where all of those biases lie. It’s where all your blind spots are. And what great insights people do, what great researchers do when you’re next to them all the time, is they’ll expose you.” The best PMs lead with product sense and validate with data — forming a conviction about what to build, then designing the experiment to test the riskiest assumptions. They do not wait for data to tell them what to build, but they also do not ignore data that contradicts their intuition.
Baxley adds a practical distinction: “I’ve never seen a product be successful that used metrics as a driver for what they were doing. I’ve seen a lot of companies be really successful seeing metrics as a consequence and a way to evaluate the quality of their decisions.” The test of product sense is whether you have a point of view about which version is better even before the A/B test returns results. If you do not, you are relying on data as a substitute for judgment.
When to Lead with Which
| Situation | Lead With | Why |
|---|---|---|
| Exploring a new product area | Product sense | No data exists yet. Judgment and user empathy must guide early decisions. |
| Optimizing an existing flow | Data literacy | Data reveals where users struggle. Product sense then proposes solutions. |
| Evaluating a strategic bet | Product sense | Strategic decisions involve too many variables for data to resolve. |
| Choosing between two specific designs | Data literacy | A/B test the designs and let users vote. |
| Responding to a competitor move | Product sense | No time for extensive research. Pattern recognition drives the response. |
| Pricing and packaging decisions | Both equally | Data sizes the opportunity; product sense shapes the structure. |
Key Takeaway
- Product sense is primarily a learned skill, not an innate gift. It develops through exposure, reflection, and deliberate practice. Treat it as trainable, because it is.
- Product sense has four components: user empathy, domain knowledge, creative ability, and technical understanding. Identify your weakest component and practice it deliberately.
- The core mechanism is pattern recognition. PMs who have analyzed many products, studied many case studies, and reflected on many decisions develop an internal library of patterns that accelerates future judgment.
- Product sense and data literacy are complementary but different muscles. The best PMs lead with product sense (forming convictions) and validate with data (testing assumptions). Neither alone is sufficient.
- Practice product teardowns, decision journaling, and cross-domain study consistently. Fifteen minutes of deliberate analysis per day compounds over years into dramatically stronger product judgment.
Related
- Roadmap Prioritization — Product sense is the judgment layer that sits above prioritization frameworks
- User Research — Research develops the user empathy component of product sense
- North Star Metric — Choosing the right NSM requires product sense, not just data analysis
- Product-Market Fit — Recognizing PMF (or its absence) is a product sense judgment call
- B Testing — Data literacy complements product sense; knowing when to test vs. trust intuition is itself a product sense decision
Sources
- Shreyas Doshi on the art of product management — Taste beyond pixels, evaluating beliefs and ideas, alliteration bias, two-way door trap, questions product leaders should ask themselves sooner
- Judd Antin on the UX research reckoning — Intuition as pattern matching, cognitive biases in the gut, falsification over validation, user-centered performance
- Bob Baxley on 35 years of product design wisdom — Software as a medium, developing taste through observing users, metrics as consequence not driver, moral obligation of product craft
- Mihika Kapoor on building zero to one at Figma — Product sense as intuition, building conviction through user conversations, Meta’s two-capability PM framework
- Dylan Field on intuition, simplicity, and the future of design — Intuition in product decisions, FigJam differentiation through fun, value moving up the stack