A Beginner's Guide to ICP for SaaS Founders

You finished the product. You set up the landing page. You ran an ad or two. And then nothing happened. Not a flood of signups, not even a trickle. You start questioning the product. In most cases, the product is not the problem. The problem is that you are trying to talk to everyone, which means you are actually talking to no one.

This is the ICP problem. It quietly affects almost every early-stage SaaS founder I have worked with. In this edition, I want to cover what an ICP is, how to build one before you have any customers, the most common mistakes, and one exercise to get you started this week.

What Is an ICP?

ICP stands for Ideal Customer Profile. It is simply a description of the type of person or company who gets the most value from your product, is most likely to pay for it, and is most likely to stick around long term. It is not a wish list of big clients you hope to land one day. It is an honest answer to: who can I actually help right now?

One thing worth clearing up: ICP is not the same as a buyer persona. Your ICP describes the type of customer, their situation, and their problem. A buyer persona describes the specific person inside that company, their role, what they care about, what puts them off. You need the ICP first. Without it, the persona has no grounding.

Here is why ICP matters for everything you do in marketing:

• Homepage copy:You cannot write a headline that lands without knowing who will read it. I covered how to turn that clarity into the right message in How To Offer Benefits, NOT Features.
• Ad targeting:Broad targeting burns budget fast. As I wrote in Why Ads Alone Won't Skyrocket Your SaaS Sales, ads without a clear audience are just paying to watch people walk out the door.
• Onboarding emails:When you know what your user signed up to achieve, you can speak to that directly. See How to Write SaaS Onboarding Emails That Convert for how to build that sequence.
• Testing:Every headline or CTA you test means more when you know who you are testing it on. More on that in A/B Testing for SaaS Marketing.

I have worked on products where the team skipped this step and went straight to paid ads. The clicks came in. The conversions did not. The landing page said something like "the all-in-one platform for growing teams," which sounds professional and means nothing to the person reading it. Once we got clear on who we were actually talking to, the copy changed, the channels changed, and the results followed.

How to Build a Starter ICP When You Have No Customers

Most ICP guides tell you to look at your best customers and find patterns. That is solid advice once you have customers. Before that, you need to start with a specific and honest guess, and then test it.

Start with the problem, not the product

Go back to the reason you built this. What specific problem were you trying to solve? Who is frustrated by that problem enough to pay someone to fix it? Write your first draft in one sentence: "People who struggle with [problem] and are currently dealing with it by [their workaround], who have a reason to switch."

Think about the situation, not just the type of person

"Marketing managers" is not an ICP. It is a job title. A marketing manager at a five-person startup running their first campaign has completely different needs than one at a 300-person company with an established team. Think about TAM and SOM here too: as I explained in How to Find the Right Market for Your SaaS, the customers you can realistically reach right now are a much smaller slice than the full market. Your ICP should reflect that slice, not the whole pie.

Find information before you have users

The signal you need exists before your first customer signs up. Here is where to look:

• Reviews on G2 or Capterra for tools similar to yours. The words people use in reviews are the exact words your future customers use to describe their pain.
• Reddit threads about the problem your product solves. Look at who is asking, what has frustrated them, and what they have already tried.
• Job listings. A company posting for three customer success roles is telling you they are growing fast and feeling the pressure.
• Short conversations. Five to ten calls with people who match your guess will teach you more than weeks of research. This is also where you find the early users I covered in The Best (and Worst) Ways to Get Your First 100 Customers.

Write it down on one page

Your first ICP draft should cover five things: who you are targeting and what their situation looks like, what problem they are trying to solve, how they are dealing with it today and why that is not working, and what a good outcome looks like for them six months after using your product. One paragraph per question. Specific, not polished.

The Five Most Common ICP Mistakes

• Trying to appeal to everyone. When you write for everyone, you move no one. Being specific about who you help is not a limitation. It is how you stand out when you cannot outspend the competition.
• Chasing the customer you wish you had. Every founder wants the big client with the impressive logo. But that client might need features you have not built yet or a level of support you cannot provide. Sell to the customer you can actually win today and build toward the bigger ones.
• Mistaking early buyers for your ideal customer. Whoever signs up in the first few weeks is not automatically your ideal customer. Some will churn quickly. As I covered in What's a Good Churn Rate & How to Improve Yours, customers who are a poor fit are one of the most common and avoidable sources of churn. Build your ICP from the people who get real value, stay, and refer others.
• Writing it once and forgetting it. Your product changes, the market changes, and the types of customers who need you most will shift over time. An ICP that was accurate at launch can quietly point you in the wrong direction a year later. Check it every few months.
• Defining the customer only by basic details. Knowing your ideal customer is a "B2B SaaS company with 20 to 50 employees" is a start, but it does not tell you much on its own. What situation are they in? Are they growing fast? Did something recently change in their business? The circumstances around the customer tell you far more than the category they belong to.

Practice: Your One-Page ICP Draft in 30 Minutes

Open a blank document and write honest answers to these five questions. One short paragraph each. Do not aim for perfect, aim for specific.

• Who is your product built for? Describe the type of company or person, their size, and what their day looks like.
• What problem are they trying to solve right now, and how are they currently handling it?
• What needs to be happening in their life or business for them to actually buy something like this today?
• Who decides to buy? Is it the founder, a manager, a team lead?
• What does a win look like for them? What changes in their work six months after they start using your product?

Once you have written your draft, find one person who matches your description and share it with them. Ask if it sounds accurate. If it does, you have a working starting point. If it does not, that conversation just saved you weeks of talking to the wrong people.

The Bottom Line

You cannot write a good headline without knowing who will read it. You cannot target an ad without knowing who should see it. You cannot write onboarding emails without knowing what your user signed up to achieve. All of it starts here.

Founders who skip the ICP step do not avoid the problem. They just deal with it later, after months of flat results and rewritten copy. A rough draft you can start testing this week is worth more than a perfect profile you never get around to building.