Why your Product Sense answer doesn't feel like a slam dunk at the Director and VP level

You can give a Product Sense answer that is good by every measure you can point to. You clarified the prompt, picked a user, found a real pain, and came up with a solution that was sensible. Sometimes the feedback still comes back that it was not quite at the level, or that you could be sharper or more strategic, and no one can tell you exactly what was missing.

That is what makes this feedback hard to fix.

An interviewer is not grading your answer point by point while you talk. Some interviewers will push on your logic and some will not, so you cannot count on getting a question that exposes the gap for you. What they are all doing is following your thinking and forming an impression as you go.

When each part of your answer leads into the next, the whole thing makes sense as one piece. When one part does not fit with another, that impression never quite forms.

Earlier in your career, the gaps in a Product Sense answer were usually more obvious: no clear structure, an unclear point, a jump to a conclusion, or a decision that looked random. Those are easy to spot because someone can point to the problem.

By the time you are interviewing for senior leadership roles, your answers are more thoughtful and those obvious gaps are usually gone. What is left is often a single mismatch buried inside a lot of strong reasoning. It slips past you. The interviewer comes away with a good answer they feel less sure about than they should, but there is no obvious mistake they can point to.

You might not catch this in your own answer because you are the only person in the conversation who knows what you meant. Your intention fills in the gaps your words leave. The answer sounds complete to you and can still fall short for them.

This article is about the two gaps that do this most often, why they are hard to see in your own answer, and how to find and close them before the interview.

Why this shows up at the Director and VP level

Earlier in your career, a structured, complete answer was often enough to pass a Product Sense interview. You followed a clear structure, covered the prompt, picked a user, found the pain, and built the solution in a sensible order.

By the time you are interviewing at the Director, VP, or L7 - L9 (Meta and Google terminology for these senior leadership roles), everyone in the running can do that well. A structured, complete answer is the baseline. What sets you apart is whether your thinking is unique and clear such that the whole answer holds together end to end: whether each choice makes the next one follow, and whether the answer comes across as one connected strategic narrative rather than a set of correct but separate thoughts.

This is different from sounding strategic, which I cover in how to sound strategic in a Product Strategy interview. Sounding strategic is about how high-level you speak and how you frame the big bet. You can get that exactly right and still have your reasoning stop holding together underneath it.

This piece is about the two places where that happens most often.

The first gap: you contradict yourself without knowing it

If I told you the most common gap is that people contradict themselves, you would probably think it does not apply to you. Almost no one thinks it applies to them. Yet it is one of the most common gaps I've seen, and it tends to show up in strong answers because a richer answer has more parts, and every part has to agree with the others.

Let's go through a real-life example from a client: Help OpenAI move into healthcare.

1. They opened by setting the scene. Big retailers were buying their way into pharmacies, other AI tools were moving into health, and people were already asking search engines about their symptoms and getting back a mess of conflicting results.

2. Against that backdrop, they set a goal: the product should help people get safe, reliable, trusted answers to their medical questions, from a first worry to a second opinion.

3. Then they went to who the user is: patients, the doctors and nurses who treat them, the pharmacies, the insurers. They chose to build for patients, the largest group and the one closest to the mission.

4. Going deeper, they laid out 4 segments for patients, ordered by how far along they are:

1. 1. Someone with a casual health question who has not seen a doctor yet.

2. 2. Someone with a more serious worry who still has not seen a doctor.

3. 3. Someone seeking a second opinion, who has already had tests and a doctor's visit.

4. 4. Someone already in treatment, with tests done and monitoring in place.

They picked the first two, the people who have not seen a doctor yet, with this reasoning: that group is large, it is underserved, and helping them fits the mission.

Notice the original goal they had set, and their reasoning for picking a segment.

The goal focuses on helping people get reliable answers. But the chosen segment is where a reliable answer is hardest to give. There are no tests and records, nothing to work from but a description of symptoms, which is the most uncertain case there is.

Groups 3 and 4 come with tests and reports already done, real data to reason from. That is where a reliable answer is easier to give.

The goal they had set points one way, toward the groups where a reliable answer is possible, while the choice points the other way, toward the groups where it is hardest.

The two do not line up, and now this narrative no longer feels cohesive.

Why this is easy to miss

The two choices happen at different points in the answer.

You set the goal at the beginning. Then, a minute later, you choose the user by looking at size and need. By the time you get to that choice, you are thinking about the user in front of you. You are not going back and checking it against the goal you set earlier.

That is what can make this challenging. The goal, user, and reasons all make sense. It is only when you put the two choices next to each other that you see they are pulling in different directions.

This is also why I see it more often in thoughtful, experienced candidates. The more nuanced your answer becomes, the more decisions you are making, and the more those decisions need to line up. You can do all of the thinking correctly at each step and still lose the thread between steps.

The fix can be simple: go back and check whether the choices you made earlier still support the choice you are making now.

The second gap: a reason that does not actually support your choice

The second gap is about the reason behind a judgement call. A reason only does its job when it shows why a particular judgement call is the right call, not just an acceptable one.

Let's use the same example above.

The reasons given for choosing the first two groups were that the group is large, that it is underserved, and that it fits the mission. All three are true, but none of them answers why this is the right strategic move for the company.

Of the three reasons, underserved sounds the most like a strategic reason, but a space being underserved does not on its own mean it actually makes strategic sense. Sometimes a space is underserved because it is genuinely hard, or genuinely risky, and the companies that could serve it have looked and decided to stay out.

So underserved is not automatically a good thing. It can be a warning sign.

What would make the the judgement call land is when it can actually move the needle for the company.

To see what that is for a company like OpenAI, look at what it competes on. Its models are not the only good ones, and a rival's model is rarely more than a few months behind. When someone can switch from one AI assistant to another in a minute, the model itself is not what keeps them.

What keeps them is habit and trust: being the one they already rely on and do not think to leave.

So OpenAI's real advantage is not a single feature that a competitor could match. It is becoming the default people come back to, the one they open again without thinking about switching.

That is consumer lock-in. You reach people early, become the one they trust, and stay their default, so that when they need help again later they come back to you instead of trying a competitor.

Health makes that even stronger. In health, trust takes a long time to earn, and once someone gives it, they do not drop it easily.

Once someone has brought their worries to you and been helped, they tend to keep coming back rather than start over somewhere new. This is where the first two groups come in.

They are the front door of a person's health journey: the moment someone first types a worry into a search box and gets back a pile of conflicting, frightening results.

Be the calm, reliable answer they find at that moment, and you earn their trust before anyone else does. Then when they reach the second opinion and the treatment later, you are already the one they come back to.

Win the front door and you hold onto them through everything that follows. That is an advantage a competitor cannot easily copy.

Consumer lock-in would be a more strategic reason for picking the first two groups compared to how big and underserved this group might be. It is also how serving them serves the mission: not simply because there are a lot of them, but because it is how OpenAI becomes the place people come to for health and stay.

The same logic holds well beyond OpenAI. Consumer lock-in matters most whenever a product is easy to switch away from and the winner is whoever becomes the trusted habit.

However, winning these two groups is harder and costs more than serving groups 3 and 4. At the very start there are no tests and no records, so making the answer reliable is a very hard problem to solve. It takes heavy investment in the model and in medical data that is hard to get.

A worried person typing symptoms into a search box also has no way to know your answer is more trustworthy than the free one they already get. You have to earn that belief before they will switch, which makes each new user slower and more expensive to win.

The bet is that the reward, becoming the default people come back to for the rest of their health journey, is worth paying that price up front.

When you go back over your own answer and find a reason that is true but only describes the option you picked, without showing why picking it is the right judgement call for this company, you have found this gap.

The fix is to actually identify the thing that would make it the right move for the company, the lasting advantage it gives the company or the mission it serves, and then show how your pick is the one that delivers it.

Identifying why a company should invest in something is part of product strategy questions as well. I invite you to go through that in my product strategy post.

Why it's hard to see either gap in your own answer

Both gaps come from the same place. They are genuinely hard to catch in your own answer, and it has nothing to do with how good you are.

You are the only one in the conversation who knows what you meant. Everyone else has only your words. You have your words plus everything you meant by them, so the parts connect in your head whether or not they connect out loud.

Also, your whole career likely has trained you to check whether something is true, and you are very good at it.

So when every statement in your answer is true, it feels finished. But whether a statement is true is a different question from whether two parts of your answer agree, and different again from whether a reason actually supports the company strategically.

Everything can be true and the answer can still have a gap. That is the check that many of my clients find challenging, because nothing ever feels that wrong. But you can certainly practice to get better at this, which is how clear thinking is applied in an interview setting.

Go back over your own answer before your next practice round

Take one Product Sense answer you have already given and go back over it slowly, twice. You are looking for the two gaps above.

1. The two choices that do not agree

Set your choices next to each other, especially ones from different moments in the answer: the goal you set and the group your solution actually serves, the user you picked and the behavior your solution needs from them, the thing you called most important and the thing you landed on.

Wherever two do not fit, change one so they do.

2. The reason that does not support the judgement call

For each reason you gave, ask whether it only describes your decision without showing why picking it is the right strategic move for the company.

If so, replace it with the reason that it can strategically move the needle for the company.

You cannot do this at full speed from the start, which is why clear thinking takes time. You do it slowly, on your own, until you start to catch the gaps yourself while you are still talking. That is when it helps you most.

A note from coaching

Our clients are already excellent. They are not missing anything.

Again and again, the moment that lifts an answer is not a new idea. It is the moment they slow down, go back over their own reasoning, and find the two choices that did not agree, or the one reason that was not actually strategic.

Over the last 12 to 18 months, our clients have landed offers at companies like Google, OpenAI, Meta, Microsoft, and Stripe with total compensation packages up to $2.9 million. The common thread in the prep was rarely a new idea. It was learning to go back over their own reasoning to get faster at clearer thinking.

That is the level of thinking I want you to practice. Your goal is not to make your answer sound more strategic on the surface, but to make sure the thinking underneath it is based on sound judgement.

FAQ

Why doesn't my product sense answer feel like a slam dunk even when it is complete?

Because at the Director and VP level (or L7-L9 using Meta/Google terminology), a structured, complete answer is the baseline. What sets answers apart is whether the whole thing works as one connected strategic narrative. The interviewer forms an impression as you talk, and when one part does not fit another, or a choice is not supported by a strategic reason, the answer will not feel convincing and they come away less sure, often with no specific note.

How do interviewers actually judge a Product Sense answer at this level?

Less deliberately than it feels. Some push on your logic and some do not, but all of them are following along and forming an impression of whether the answer holds together. A gap may not register as a mistake they caught. It can register instead as a moment that does not quite flow, which is why the feedback you get is often vague.

How do I find the gaps in my own answer before the interview?

Go back over it slowly, one point at a time. First, set your choices next to each other and look for any two that point different ways, like a goal you set and a segment you then picked for reasons that work against it. Then, for each reason, ask whether it only describes your decision or whether it actually shows why picking it is the right strategic move. You are looking for the parts you could not defend if someone asked you to explain them, because that is where the answer will feel less convincing.

Why can't I see these gaps myself?

Because you are the only one who knows what you meant, so your intention fills the gaps your words leave, and the answer feels complete to you. You are also used to checking whether each statement is true, and every statement can be true while the answer still has a gap because being true and being strategic are different things. Going over it slowly, the way a stranger would hear it, is what brings them out.

Is Product Sense at the Director and VP level really different, or is it the same interview?

Same prompts, higher bar. Earlier, a structured, complete answer was enough to pass. At the Director, VP, or L7 - L9 using Meta and Google terminology, everyone's answer is structured and complete, so what separates people is whether the argument holds together the whole way through. The same walk-through with nothing out of place beats the same walk-through with a gap in it.

Where to start

If you want another set of eyes on your own answers, that is the core of what I do with product leaders preparing for Director, VP, (or L7 - L9 using Meta and Google terminology) interviews. You can apply here, or apply here for all other roles. Real client outcomes are at nancychu.co/wins.

About the author. Nancy Chu is an L7 - L9 PM leadership coach and a former Director at Roku and manager of product managers at Meta. She works with product leaders preparing for interviews at companies like Google, Meta, and MAI, on clear thinking, product sense, and leadership presence. Her clients have landed offers with total compensation packages up to $2.9 million.

More clear-thinking insights. Every week I break down the exact thinking that makes an answer convincing, the gaps I catch and the moves I use with clients. You can get them each week.