How to answer a product strategy question like "How would you 10x Google Gemini" at the Director and VP level

To answer a product strategy question like “Grow Google Gemini” / "10X Google Gemini" at the GPM, Director, or VP level, you reimagine the product instead of improving it. Improving it means making the same product incrementally better, which competes with its immediate rivals and keeps you at a 2x answer. Reimagining it means asking what the user is really there to do, and competing with everything else people use to meet that same or similar need, which is where a 10x answer lives.

This article describes the thinking that gets you from 2x to 10x: how you reframe what the product is competing for, how you reason your way to more than one bet that only this company could make, and how you choose the one to commit to.

Reimagining a product this way is clear thinking applied to a product strategy question, and it is the difference between an answer that sounds structured but lukewarm and one that sounds like it came from someone who is a strategic leader.

Why reimagining a product is the work at the GPM, Director, and VP level

Reimagining a product is not something you switch on for interviews. It is how the job actually works at this level. At the GPM, Director, and VP level, part of the work is to set a bold direction for a product and bring people with you, and a strategy question is one of the moments where you have to do that on the spot.

You may already think this way in your own work without ever working out how you do it. That holds up until a moment asks for it fast, like an interview, a board review, a podcast, or a hallway question from your CEO. When the skill lives only as instinct, it can go missing right when the pressure is on. Working out what you already do, so you can see it and use it on purpose, is how instinct becomes clear thinking you can produce on demand.

This is worth practicing even when you are already good at it. A client we are working with was asked by their CPO to think bigger, drop the usual constraints, and bring back a bold proposal. The instinct was already there. What we did was make the thinking explicit, so they could use it deliberately for any moment, instead of hoping it showed up.

At the GPM, Director, and VP level (Meta and Google call these L7 to L9, and at most other companies they map to Director and above, which you can check on levels.fyi), a clean, well-structured answer is something every serious candidate can produce. What can separate 2 strong candidates is the thinking only one of them brings, the reimagining that a thorough answer does not reach on its own.

Google says as much about what it looks at. Its own candidate guidance for the PM interview explains that the questions are open-ended on purpose, so the interviewer can see how you solve problems, and that they pay more attention to your thought process than to the answer itself. A polished 2x answer is one of the most common ways a strong candidate lands a level below the one they were targeting.

2x thinking versus 10x thinking, and how to get from one to the other

2x thinking improves the product you already have. It makes the same experience better, and it competes with the product's immediate competitors, the products it goes head to head with today.

10x thinking reimagines the product. It changes what the problem is, or what the constraints are, so the experience becomes something new. It competes with the product's adjacent competitors, the other things people turn to for the same or similar underlying needs.

To get from 2x to 10x, you can map the competition in 2 categories:

1. Start with the product you are asked about.

2. List its immediate competitors, the products it goes head to head with today. Competing with them keeps you at a 2x answer.

3. List its adjacent competitors, everything else people use to meet the same or similar underlying need. Competing with them is what moves you toward 10x.

4. Build your answer toward the adjacent competitors.

I like using the Product-Market Expansion Grid to help me visualize this mental model. One axis is the market, either the one you already serve or a new one. The other axis is the product, either the one you already have or a new one.

A better product for the market you already serve sits in the 2x corner, and a new product for a new market sits in the 10x corner. The 2 corners in between, a new product for your existing market and your existing product taken into a new market, are steps along the way.

Take YouTube for example. Its immediate competitors are the other places you watch videos, so a 2x answer makes video better, with sharper recommendations, less buffering, and cleaner search. To reach a 10x answer, think about what YouTube is really competing for beyond just being a video app. It is competing for your free hours against everything else you could spend them on, whether that is eating, sleeping, seeing friends, learning, or going out.

Once you see it that way, the biggest version of YouTube is no longer about video at all and becomes about owning that free time out in the world, not just on the screen. That is the leap Disney made. Disney did not set out to only make better cartoons than other studios. It went after your entertainment and built theme parks and an entire franchise toward that 10x mission.

That is an example of new product for a new market, and it is a different size of ambition.

How would you 10x Google Gemini?

Now let's work through the prompt on 10x’ing Gemini, using the same thinking.

I approach every strategy answer the same way: Lean into the company's biggest strength, address its biggest weakness, and keep delighting users in ways that are hard for competitors to copy. In the short term compete with the immediate competitors, which is how we can 2x. In the long term, compete with the adjacent competitors, which is where 10x lives.

Google's strength is the ecosystem no rival has: Search, Android, Chrome, Gmail, Docs, Maps, Photos, and Workspace, with the Google account that ties them into one identity. As of the January 2026 deal where Apple licensed a custom Gemini model to power the rebuilt Siri, Google's model reaches the iPhone too. That is distribution, and a set of places to show up, that a competitor cannot simply buy nor copy.

The weakness used to be the model, but not anymore. Gemini has caught up on quality and sits as a strong number 2. The category leader consumers open first is still ChatGPT. What Gemini is winning is the race to become the assistant you open by default, and it is winning that on Google's own distribution, the Android phones, the Chrome browser, and the Search bar that already reach billions of people.

Once the models are about equally good, being the assistant more people open by default is not a lasting advantage, because the moment a rival is just as good and just as easy to reach, there is nothing keeping people with Gemini. And that is the weakness we want to minimize.

Here is an example of a 2x answer:

1. Lay out Gemini's strengths and weaknesses.

2. List the competitors, which are the other chat assistants like ChatGPT, Claude, and Copilot.

3. Split the users into enterprise and consumer. Pick the consumer segment which fits their core ads business.

4. List pain points for consumers: Gemini can’t do much beyond chatting, it’s not integrated deeply with Gmail, Google Pay, Youtube, etc.

5. Come up with the improvements: a stronger model, a cleaner interface, deeper Search built in, better memory of you and your work, an assistant that beats ChatGPT on the everyday things people already use an assistant for.

6. To go bigger, add the ideas where Gemini replaces the human assistant you use for basic admin, to pay, or the tutor, or the travel agent, etc.

The reason why this is 2x is because the improvements make Gemini a better version of what it already is, by competing with immediate competitors, so the whole answer stays measured against ChatGPT. The human assistant replacement ideas that move your calendar or plan your trip is what ChatGPT is racing to build too. When your closest competitor is chasing the same target, that target is one of your immediate competitors, not an adjacent one. Replacing a person in software feels like reinvention and reimagination but is still an improvement on the assistant you already have.

Mapping out the strengths, competitors, users and painpoints is a different task from reimagining the product, and you can finish all of the mapping and still not have reimagined anything, because laying out the pieces does not do that reimagination for you. This is usually where a 2x answer ends.

Here’s how we can reframe and think toward 10x.

Google's mission is to organize the world's information and make it useful. A chat assistant does the first half of that, it organizes information and hands it to you. The fuller version of the mission is to act on that information out in your world, so your day moves forward without you having to stop and go get it. So instead of thinking about how to build a better assistant than ChatGPT, think about what a Gemini user is actually there for: They are there to get through the day and to do it with more ease, ie without stopping to summon a tool every time.

That is the underlying need for humanity. Now look at what people already use to meet that need, those things are the adjacent competitors for Gemini. People get through the day on the phone in their hand, the car on the drive, Maps on the way, the camera pointed at whatever is in front of them, their own memory and attention keeping it all straight, and, for a lucky few, a human chief of staff who anticipates what they need before they ask. Not one of those is another chatbot.

This is the shift, from competing with the other assistants to competing with everything people already lean on to get through the day, and it is what opens up a 10x answer.

Once you have made that shift, you can envision more than one 10x direction. The technique is repeatable. Take a piece of that underlying need around getting through the day, and pair it with the Google asset a rival would find hardest to copy. Each pairing will point you to a different 10x idea.

The first pairing takes acting in the world and the surfaces Google already sits on, the phone, the car, Maps, and the camera. You get an ambient Gemini, one that acts in the place you are instead of in a browser tab or an app. Imagine:

• You're driving and it sees the meeting moved, reroutes you in the car's own display, and pays the toll and books the parking through Android Auto.

• You walk into your house and it has already turned on the lights, set the temperature, and started the playlist off your Calendar and your patterns, through the Nest devices in the walls.

• It reads the screen on any app on your phone you're in and acts on it without you switching to a chatbot, because it's built into Android.

What it competes with is the patchwork you use to get through the day now, the separate apps, the smart-home gadgets, the map, and the notes you leave yourself, not another assistant in another app.

The second pairing takes anticipating your life and the signal only Google holds across Search, Maps, Gmail, Calendar, location, and your search history. You get a proactive Gemini, one that acts before you ask. Imagine:

• It sees the flight delay and rebooks the connection.

• It notices the meeting moved across town and so it tells you to leave earlier.

• It sees your shopping searches and advises you on blind spots you might have as you make your purchase decision.

What it competes with is your own attention and the reactive tools you juggle today, the ones that wait for you to open them and ask.

These are different 10x directions for Gemini, an ambient one that acts in your world and a proactive one that gets ahead of you, and both sit with the adjacent competitors, because both compete with how people get through the day, not with the next chatbot.

You decide the adjacent competitors from what the user is there to do beyond what they use chatbots for today, the same way YouTube's adjacent competitors come from what people do with their free time, beyond what they use Youtube for today.

How to choose which 10x idea to commit to

There are 2 important signals in product leadership candidates that you need to showcase across various interview formats: vision and judgement. Product strategy prompts are great at showcasing them.

You now have multiple 10x directions for Gemini, the ambient one that acts in your world and the proactive one that gets ahead of you. Coming up with more than one bold direction that only Google could take is the vision half of the answer. Choosing between them is the judgment half, and the judgment half is where your seniority level can really surface.

Start by throwing out any idea that does not feed the business. A bold idea that does not strengthen the company you are building for will not make strategic sense for that company. Purely reaching for the “biggest reach” also will not make strategic sense. What will make strategic sense is how tightly the idea connects to the business the company already depends on.

My mental model is based on how an investor decides whether something is an a worthy investment by thinking through these 4 dimensions:

1. Does it fit the mission the company is built around?

2. Does it strengthen the core business that the company is based on?

3. Does it build on a strength the company already has?

4. Is there real signal that people want it, in a way that will have a positive flywheel effect on the core business?

Let’s go through a counterexample first. Gemini launching a social app to beat Meta AI is a big market with no line back to Google's strength, so it fails the investment check.

Now let’s take our 10x ideas on an ambient and proactive Gemini through these 4 dimensions. They both pass. Each fits the mission of organizing the world's information and making it useful, each defends Search and the advertising behind it, each is built on an asset only Google has, and the demand is already there, because people want their life handled with a lot more ease.

When more than one idea clears the investment bar, I go back to the one-liner we started with:

1. Does it lean into the asset only Google has, since our strategy is building on a strength no competitor can easily copy.

2. Which idea most protects the core business Google is defending right now, which is Search and the advertising behind it.

The ambient idea leans on the phone software, the car, and Maps, the places Google already sits that a competitor cannot simply copy.

For the proactive idea, the exclusive asset is the signal Google holds across Search, Maps, Gmail, Calendar, and location at once, the one combination no competitor has put together.

Both pass, so onto the second question.

Anticipating what you need before you ask is the same intent that Search and advertising are built on, so the proactive bet feeds the core most directly. Getting ahead of you on the flight, the meeting, and the purchase decision deepens exactly the signal the ad business depends on. The ambient bet is close behind and defends the core a different way, by making Google's own surfaces the reason to stay, so it is the strong second.

Therefore, proactive Gemini is the 10x bet I would prioritize first.

Showing the judgment part is not just landing on the prioritization, but actually mapping out your reasoning that make good strategic sense for the company, and not basing it on mechanical guesses like “reach”, “level of pain”, “willingness to pay”. When an idea does not pass the investment bar, you’d filter those out; and the remaining ideas you’d identify the one that feeds the core business the most.

Talking through the choice as a piece of reasoning, and holding it when it is pushed on, is what separates a deliberate decision from a random guess.

What to practice this week

Take one product, start by reframing.

Instead of thinking about how to improve the product, think about what its user is really there to do, and list everything else people use to do that same or similar job. Those are the adjacent competitors, and they are where a 10x answer comes from. Then generate more than one bet by pairing each of those jobs with the strength this company has that a rival would find hardest to copy. Each pairing can be a 10x candidate. Aim for multiple, so you can practice surfacing the vision signal, identifying more than one bold direction only that company could take.

Then practice the judgment half. Throw out any candidate that does not feed the business and doesn’t meet the investment bar, whether it fits the mission, strengthens the core, builds on a real strength, and shows demand that feeds the business. And also think through the least copyable asset and the tightest fit to the core, and prioritize that top idea.

The mechanics are the easy part. The hard part is committing before you feel certain, and holding your idea with conviction when the interviewer pushes on it or pivot if makes more sense to. Staying in analysis until you run out of time, or using a tactical reasons like the “reach is bigger” would be common reasons why the judgement signal is weak.

FAQ

What is 2x versus 10x thinking in a product strategy interview?

2x thinking improves the product you have, making the same experience faster or cleaner, and competes with its immediate competitors. 10x thinking reimagines the product, increases the problem space, or the constraints, and competes with its adjacent competitors, the other things people turn to for the same or similar need. At the GPM, Director, and VP level, a thorough 2x answer is something every serious candidate can produce, so the reimagining is often what separates a strong answer from an ordinary one.

What is the Product-Market Expansion Grid?

It is a simple picture of where a 10x idea lives. One line is the market, either the one you already serve or a new one. The other line is the product, either the one you already have or a new one. A better product for the market you already serve sits in the 2x corner, and a new product for a new market sits in the 10x corner. The 2 corners in between are steps along the way to 10x.

How do you come up with more than one 10x idea in a strategy interview?

Start by reframing the question, from how to improve the product to what its user is really there to do. List the other things people use to do that same or similar job, which are the product's adjacent competitors. Then pair each one with the strength the company has that a rival would find hardest to copy, and each pairing points to a different 10x bet only that company could make. For Gemini, pairing acting in the world with the phone, car, Maps, and camera gives an ambient assistant, and pairing anticipating your life with Google's cross-product signal gives a proactive assistant that gets ahead of you.

How do you choose which 10x idea to commit to?

First throw out any idea that does not feed the business, using the 4-question check on whether this is a good investment idea: does the idea fit the mission, strengthen the core, build on a real strength, and show real demand. Among the ideas that pass, break the tie as needed: which leans hardest on the asset only that company has, and which most protects the core business it is defending now. Describing that reasoning rather than only landing on an idea is what makes it a deliberate decision rather than a random guess.

Why does a 10x idea have to strengthen the core business?

Because a bold idea that does not strengthen the business would not make strategic sense for the company. A bet earns the word strategic by how tightly it connects to the business the company already depends on. Before you commit to any 10x direction, check it the way an investor separates an investment from an expense: does it fit the mission, strengthen the core that makes the money, build on a real strength, and show demand that comes back to the business. A huge but disconnected market, like Gemini launching a social app to beat Meta AI, fails that check no matter how big the reach is.

How is a product strategy interview different from product sense?

A product strategy answer ends at a high level, the direction the company should take and the bet the company should make to 10x and not just 2x. A product sense answer goes further, into the specific users, their pain points, and the solution you would build.

Want to work together?

If you are preparing for a senior product leadership interview, that is the work we do together.

For senior leadership roles outside of product, click here.

You can see recent client wins at nancychu.co/wins.

For how leveling actually works at the top, read the companion piece on the L7 to L9 product leadership interview. And for the thinking underneath a strong strategy answer, how to find the crux of a problem and decide with incomplete information, read Product judgment: how to decide with incomplete information.

About the author

Nancy Chu is an L7-L9 PM product leadership coach. She was a director at Roku and a manager of product managers at Meta before coaching full time. She works with senior product leaders at companies like Google, Meta, and Microsoft, helping them land roles at $900K and above in total comp and lead with more presence once they are in the seat.

Every week she breaks down the distinctions that matter in senior leadership product interviews, like how to reimagine a product for a 10x answer and what separates an L7 answer from an L8 one. You can get them each week here.