How to Define Your ICP Before Scaling
Your ICP Is Too Broad to Scale (And Here's How to Tell)#
Your first five customers probably arrived by accident.
One came from a tweet. One from a warm intro. One found your landing page while searching for something adjacent. You closed them all, and that felt like proof.
It was. Just not the proof you think.
The Moment the Vagueness Becomes a Problem#
At validation stage, messy customer acquisition is fine. You are learning. Every close teaches you something.
At scale stage, that same messiness is expensive.
The moment you try to hire a demand-gen lead or launch a paid channel, someone asks: who do we target? And if your answer is "mid-market SaaS" or "companies with this pain point" or "teams under 50 people," you are about to waste money finding out the hard way that none of those are targeting definitions.
They are market descriptions. Those are not the same thing.
What an Actionable ICP Actually Does#
An ICP is not a demographic summary. It is a filter.
It tells you which company to call tomorrow. It tells you which company to skip. If your definition is so broad that half your addressable market fits it, you do not have an ICP. You have a hypothesis.
This matters because every downstream decision at scale runs through it.
Your demand-gen lead uses ICP to pick ABM accounts. Your paid ads use ICP to build audiences. Your cold outreach uses ICP to build lists. Your sales team uses ICP to qualify inbound. If the definition is vague, all of those bets are guesses. They will find the real pattern eventually, but they will burn through budget and goodwill doing it.
You are not behind for not knowing this. Most founders who have PMF and early revenue believe that having happy customers means they know who their customer is. Those are different things. You have closed deals. You do not yet have a narrow enough profile to act on.
How to Narrow It Down Before You Spend Anything#
Stop guessing. Start with your ten best closed deals.
Not your ten biggest. Your ten most profitable, smoothest, most likely to renew, most likely to expand. The ones that felt almost easy. The ones where the problem was obvious, the decision moved fast, and the customer never questioned the value.
For each deal, pull five attributes.
Three firmographic attributes: industry vertical, company size by headcount or revenue, and go-to-market motion or business model.
Two situational attributes: what specific problem they were trying to solve when they found you, and what constraint or timeline was pushing them to solve it right now rather than next quarter.
Now look at all ten deals together. Which three firmographic attributes appear in at least eight of them. Which two situational attributes.
That intersection is your actionable ICP.
Not "mid-market SaaS." Something like: Series B HR tech companies with 50 to 200 employees, where a new CFO has set a cost reduction target and hiring time is the obvious lever.
That is narrow enough to act on. A demand-gen person can find those companies. A paid channel can reach them. A cold outreach list can be built from them. A sales rep can qualify an inbound lead against them in under two minutes.
What to Do If Your Best Deals Don't Look Alike#
If you run this exercise and your ten best deals look like ten different companies, your product does not have a narrow use case yet.
This is not fatal. It means you have a horizontal tool that solves adjacent problems for different buyer types. That is a real thing. Some great businesses start this way.
But it does mean you cannot use paid acquisition or ABM effectively yet. You stay on inbound, warm intros, and direct sales until one segment pulls ahead. You watch your data. You let the pattern emerge rather than forcing one that does not exist.
Some founders try to solve this by picking an ICP anyway. That does not work. You cannot narrow to a definition your customer base does not support. Your job is to find the segment that is already overrepresented in your best deals and bet on it, not to invent one that sounds clean on a slide.
This is also where your sales process working by accident becomes a real ceiling. If you have not diagnosed why your best customers stuck around, you cannot build a repeatable pattern from them.
The Test Your ICP Has to Pass#
Before you hire a marketing lead or brief a paid channel, run this check.
Read your ICP definition out loud. Then ask: can I use this to decide whether a specific company is worth calling next Monday?
If the answer is yes, you are ready.
If you hesitate, or if the answer depends on a conversation, or if you are not sure, the ICP is still too broad. Go back to the ten deals. Look harder for the pattern.
An ICP earns its name by narrowing your options. If it does not do that, it is not helping you. It is hiding the problem.
The founders who scale well are not the ones who move fastest. They are the ones who got precise before they got loud. Precision first. Then channels. Then headcount.
If you have the closed deals but not the pattern yet, Nexdation works through this exact exercise with founders before anything else gets built.