What Does It Mean to Have “Product Sense”? (And How to Build It)
Ask 10 great product leaders to define product sense and you’ll hear variations on the same theme: it’s the ability to make consistently good product decisions in ambiguous situations. Shreyas Doshi puts it crisply: product sense is the ability to usually make correct product decisions, both macro and micro, even in the face of major ambiguity. (Maven)
That sounds like intuition-and in many ways it is. But the best people will also tell you it’s not magic. Marty Cagan argues that strong product sense is really deep product knowledge earned by immersing yourself in a domain, customers, and the competitive set. In other words, instincts get trained. (Silicon Valley Product Group)
This article unpacks what product sense is, the ingredients that make it work, and practical ways to develop and demonstrate it.
A working definition
A useful way to define product sense is as a repeatable decision‑making skill that blends:
Empathy (for users’ contexts, constraints, and motivations).
Causal reasoning (clear hypotheses about how a change will create value).
Market awareness (the competitive and business realities your product operates in).
Taste and craft (the ability to recognize and shape high‑quality experiences).
Evidence discipline (knowing when to lean on data, when to run an experiment, and when to ship a well‑reasoned bet).
Put differently: product sense is the judgment to choose the right problem, the right solution shape, and the right next learning step.
It is fundamentally customer‑first. Steve Jobs captured the mindset: “You’ve got to start with the customer experience and work backwards to the technology.” (All About Steve Jobs)
And it is anchored in markets. Marc Andreessen’s concise definition of product/market fit-“being in a good market with a product that can satisfy that market”-reminds us that taste without market truth is just opinion. (Pmarchive)
Why product sense matters (and is measurable)
Good product sense correlates with business results because it improves what you choose to build and how you validate it. Consider two data points:
McKinsey’s multi‑year study found companies in the top quartile of their McKinsey Design Index grew revenues 32 percentage points and total returns to shareholders 56 percentage points faster than peers-evidence that superior product quality and customer experience compound. (McKinsey & Company)
At Microsoft and elsewhere, large‑scale experimentation shows how often our instincts are wrong. In one HBR summary: roughly a third of ideas improve metrics, a third are neutral, and a third make things worse-a humbling reminder that product sense shines when it guides what to test next, not when it claims infallibility. (Harvard Business Review)
Together, these suggest a practical definition: product sense = user‑centric hypotheses + disciplined learning loops.
The five building blocks of product sense
1) Empathy rooted in real contexts
You can’t intuit great products from a conference room. Field studies and contextual inquiry expose the constraints and workarounds that users rarely articulate. NN/g summarizes the value: observing people in their natural environment reveals problems “you wouldn’t know to ask about,” and often changes what you design in the first place. (Nielsen Norman Group)
How to practice
Shadow customer support calls, watch screen‑share sessions, or do short field visits.
Capture moments of struggle (where, when, who’s nearby, what else is on screen).
Write one‑sentence “jobs” that describe the progress the user is trying to make.
2) Causal mental models
Product sense is not just “what users say”; it’s why the world behaves as it does. A robust approach is Jobs to Be Done: identify the job customers are “hiring” your product to do, then design around it. Christensen’s HBR piece makes the case that focusing on jobs-rather than personas or raw demographics-improves innovation hit rates. (Harvard Business Review)
How to practice
Translate observations into causal statements: If we reduce X friction in moment Y, Z adoption metric should move.
Distill hypotheses into plain, testable language.
3) Market and business literacy
Taste without market fluency is a liability. Product sense requires understanding the ecosystem your product lives in (competitors, switching costs, channel constraints) and what winning looks like (business model mechanics, unit economics). Andreessen’s PMF framing is a helpful north star for prioritization and timing. (Pmarchive)
How to practice
Keep a “market map” one‑pager: substitutes, complements, entry points, lock‑ins.
During prioritization, explicitly state the market bet you’re making and the disconfirming evidence that would change your mind.
4) Taste & craft (the quality bar)
“Quality” is not subjective hand‑waving; it’s the ability to recognize when an experience is clean, obvious, and complete. The business case is strong: better design correlates with materially better growth and shareholder returns. (McKinsey & Company)
How to practice
Build a reference library of flows you admire; do teardown notes on why they work.
Sweat “first mile, last mile” details (onboarding clarity, error states, edge cases).
Borrow constraints from the best: mobile‑first content limits, progressive disclosure, accessible defaults.
5) Evidence discipline (experiments over arguments)
Strong product sense knows when to seek fast feedback. Large‑scale A/B testing research shows many “great ideas” don’t move the needle-and some harm key metrics. That’s not an indictment of intuition; it’s a roadmap for how to learn: run small, trustworthy experiments and iterate. (Harvard Business Review)
How to practice
Decide before launch which outcomes would count as success or a rollback.
Instrument minimally but meaningfully; define leading indicators and guardrails.
Treat inconclusive results as signal about your hypothesis quality, not just sample size.
A helpful backdrop here is the Kahneman–Klein joint paper on intuition: expert “gut feel” is reliable when the environment is sufficiently regular and feedback is rapid-exactly what disciplined product teams create with continuous discovery and experimentation. (PubMed)
Concrete routines to build product sense
You don’t need a decade at a FAANG to develop product sense. You need reps and feedback loops. A practical weekly cadence might include:
Customer hours
Block two hours every week for direct contact: interviews, field observations, or watching support tickets get resolved. Field studies are especially useful in discovery because they often change what you’d build by revealing unspoken constraints. (Nielsen Norman Group)Opportunity Solution Trees (OSTs)
Borrow Teresa Torres’s OST to map from an outcome → opportunities (user needs) → candidate solutions → experiments. OSTs visualize assumptions and keep the team aligned on learning paths, not just feature lists. (Product Talk)North Star + inputs
Use Amplitude’s North Star Framework to anchor your product’s value in a single North Star Metric (NSM) and a small set of input metrics you can move. This gives your bets a quantitative backbone and reduces thrash. (Amplitude)Write the press release (or narrative) first
Before spec’ing, write the customer‑facing story: who is this for, what changes in their life, why now? It forces clarity on value and trade‑offs-then your PRD or spec can translate promises into testable requirements.Run more small experiments
Heavyweight tests aren’t necessary. Even simple A/Bs or smoke tests can deliver the rapid feedback conditions where intuition sharpens. Kohavi and Thomke’s work summarizes the power of small, frequent experiments in improving decision quality. (web-docs.stern.nyu.edu)
How to demonstrate product sense (at work and in interviews)
1) Start with the problem and the outcome.
When presented with a request (from a stakeholder or an interviewer), restate the user problem, name the North Staryou’ll move, and surface the constraints. This signals that you can separate what matters from implementation details. (Amplitude)
2) Reveal your causal chain.
Draw the simplest link from intervention → behavior change → metric movement. Then list the top assumptions that could break the chain and how you’d test them (OST language helps). (Product Talk)
3) Make trade‑offs explicit.
Great product sense is often visible in what you don’t do. State what’s out of scope for v1 and why (risk, complexity, opportunity cost). Tie those choices to your metric guardrails.
4) Show your taste with specifics.
Do a three‑minute teardown of a comparable flow (e.g., “first‑time activation”). Talk through why a particular microcopy, progressive disclosure, or error state reduces cognitive load and increases completion.
5) Close the loop with an experiment plan.
Describe how you’d instrument the experience, what constitutes success or rollback, and how quickly you’ll learn. Cite the reality that many ideas don’t work-and that your plan accounts for that with cheap, trustworthy tests. (Harvard Business Review)
Common misconceptions
“Product sense is innate.”
Cagan’s point is the antidote: it’s mostly deep product knowledge and context you can acquire with immersion and practice. (Silicon Valley Product Group)“If you talk to users, you don’t need to test.”
User research reduces unknowns, but large‑scale online experiments repeatedly show that teams overestimateimpact. Both are essential: research to choose what to build, experiments to verify that it works. (Harvard Business Review)“Taste is subjective; metrics are objective.”
Taste without metrics can drift; metrics without taste can optimize for the wrong thing (e.g., clicks over value). The North Star framework exists precisely to align craft and measurement. (Amplitude)“The market will figure itself out.”
Product sense requires market truth. PMF is about a good market and a product that satisfies it; ignoring either side undermines judgment. (Pmarchive)
A simple self‑assessment you can run this month
Evidence: Did you talk to at least five users in context about the problem you’re addressing? What did you learn that changed your plan? (If nothing changed, you likely didn’t go deep enough.) (Nielsen Norman Group)
Model: Can you write your causal hypothesis in one sentence? (If X at moment Y, then Z.)
Market: Could you explain your market map on a whiteboard in under two minutes? Where is the wedge?
Taste: Do you have a gallery of best‑in‑class flows you review with your team? What principles do they illustrate?
Learning: Do you have one experiment running (or queued) at all times? What would cause you to roll back?
If you can answer “yes” to most of these, you’re practicing product sense-not as a buzzword, but as a daily habit.
Closing thought
Strong product sense doesn’t mean you’re always right. It means you start with the customer, think clearly about cause and effect, respect the market, sweat quality, and learn fast. Jobs’s mantra-customer experience first, technology second-keeps you grounded in value. Andreessen’s PMF definition keeps you honest about markets. McKinsey’s data reminds you that craft and results are linked. And the experimentation literature keeps you humble enough to test your best ideas quickly. Put together, that’s not mysticism. It’s a practice you can build-one outcome‑oriented week at a time. (All About Steve Jobs)