Define your ICP

Who is going to buy your product?

You've probably already got a sense of that—for example, maybe you're building practice management software for dentists. But are they independent dentists or part of a chain? Dentists globally, or in a specific geo or state? Maybe you have an insight that left-handed dentists are underserved by existing solutions, and you're going to start in Iowa.

That last example might sound absurd, but it makes a point: The more specific you can be about your Ideal Customer Profile (ICP), the better your chances of building something people actually want to buy.

For an even deeper dive on the importance of a tight ICP, we recommend this resource from Notion Capital.

Avoid building technology looking for a problem

Many founders justify this by saying their product is “horizontal.” But even if true, to bring it to market you’ll have to choose use cases and an ICP anyway. People don’t buy software because it’s horizontal, they buy a specific application of it.

Can the technology-first approach work? Sure. But more businesses fail when they start with the technology rather than the buyer. If you start with the buyer and you learn everything you can about them, your likelihood of finding a problem that they care about—and building a solution to solve that problem—is much, much higher. You’ll also better understand where to find more of them to sell to later.

Defining your ICP is the foundation of everything that comes next. This decision cascades into every strategic choice you'll make: the type of seller you'll eventually hire, the go-to-market (GTM) motion that will work for you, and the channels where you'll find traction.

Get this wrong, and you'll spend months—or years—spinning your wheels talking to the wrong people about the wrong problems.

The questions you should be asking about your ICP

Do you have genuine empathy for them?

If you’re successful, you will be talking to these people every day for the next 15 years—taking support calls on the weekend, attending industry conferences, and writing content that resonates with them. If you don’t love solving problems for this group, it’s not a great idea to build a business around them.

Are they actively trying to fix the problem?

“The truth is, when somebody says, ‘Yeah, this has been a massive problem for me for the last year,’ you need to ask, ‘But why haven't you done anything about it?’” says Riya Grover, co-founder of Sequence. “‘Have you looked at any solutions in the space? Have you tried to buy anything? Have you got any budget allocated internally? Is there anyone on your team working on trying to solve this?’ And if the answer is ‘no,’ it's very often the case that it's not a high-priority problem or pain point to solve.”

Is the problem pressing?

If you’re going to close deals, there needs to be a sense of urgency that your customers need to solve the problem right away. If no one you talk to has a timeline to fix the problem, it’s not a big enough pain point for them and probably not a good fit.

You also need to consider macro issues in the industry you’re planning to sell into. For example, Sequence pivoted from a payments solution to revenue operations for SaaS finance teams because software pricing and contracts were getting much more complex. The move to AI, with outcome-based pricing and models based on token usage, made it an urgent problem.

Do they have a budget?

Your ICP needs to be able to pay you. A pressing, real problem counts for nothing if the person experiencing it can’t make the purchasing decision. If the person approving the budget doesn’t feel the pain or understand the problem, or a third party is required to implement it, your sales motion becomes much more complex.

Do you understand them?

You need to develop an intimate understanding of their workflows and the job-to-be-done for which they are hiring your product. A great example is the co-founders of ServiceTitan, a SaaS product for commercial and residential trades. Their parents were building contractors, so they had an innate understanding of the space.

Not every founder is going to have that advantage. But if you want to build for dentists, get as many cavities as you can and… just kidding. Spend the next 3 months talking to as many dentists as you possibly can.

Are there lots of people like this?

The more people who fit your ICP, the better, right? Not necessarily, but the answer is also more nuanced than "just find a niche."

The conventional wisdom is to go narrow, and there's real truth to it. A large total addressable market (TAM) can cause you to be spread so thin that your pitch doesn't land with anyone.

For Qwilr, a horizontal product that helps create sales proposals, Mark found success by starting narrow: "In the early days, I remember finding three or four niches that worked very well for us and expanding it slowly from there. We did this qualitatively and quantitatively—we spoke to lots of people to understand how they were using our product, and did some analysis to pattern match, finding industries and titles that were overrepresented."

"Market segment is very important. We kept zooming in, and every time we thought we had zoomed in enough, we never had," says Enzo Avigo, co-founder of June, which started as a product analytics tool but became a customer success platform after listening to its customers.

But there's a counterargument worth considering: Big markets are underrated. Most founders are terrified of crowded markets—so-called "red oceans"—but competing in an established market means demand is already proven and budgets already exist. You don't need to educate buyers that they have a problem; you just need a better way to solve it.

Consider the difference between Trello and Linear. Both are essentially tools for tracking work. Trello went horizontal—millions of active users, sold for over $300 million, but with less than $10 million in revenue. It didn't solve enough of the pain point for any one group to monetize effectively. Linear took a similar concept but built it specifically for software development teams. Customers love it, pay a lot for it, and the company is growing fast. Same underlying problem, very different outcomes—because Linear chose a lane within a large market.

The real lesson isn't "big TAM good" or "big TAM bad." It's that even in a massive market, you still need a clearly defined ICP. Go after boring problems that people currently solve with incumbents they hate.

Your ICP influences everything else

Your ICP will change over time, and that's fine. The key is being intentional about who you're serving right now—not trying to be everything to everyone from day one. A strong community within your niche can be more valuable than a massive TAM, particularly when you're starting out. But if you're entering a big, proven market? Find your niche within it.

Your ICP choice cascades into every other strategic decision you'll make: the type of seller you hire, the motion you choose, and the channels that will work for you. That’s why it’s not uncommon for successful startups to have gone through several rounds of picking and repicking an ICP.