Ask a founder who they sell to and you will usually get a list. Mid-market SaaS, but also agencies, plus the odd enterprise deal when one walks in, and honestly anyone with budget. Then they ask which channel to invest in, outbound or ads or content, and the honest answer is that the channel is not the problem. Most go-to-market plans stall because they are trying to sell to everyone at once.
What a fuzzy ICP actually costs
The cost never shows up as one big failure. It leaks out of every part of the motion at the same time. The message dilutes first. Copy that has to work for a fintech CFO and an agency owner ends up saying nothing sharp to either. You get a homepage full of safe abstractions and cold emails that read like they were written for a category, because they were.
Then the lists go soft. If you cannot describe exactly who you are after, every targeting filter is a guess, and the list fills with people who were never going to buy. And finally the data turns to noise. When a campaign underperforms you cannot tell whether the message missed or the audience was wrong, because the audience was five audiences wearing one name. Nothing you learn compounds, so every month starts from scratch.
A fuzzy ICP does not fail loudly. It quietly taxes every channel until nothing works well enough to scale.
How to actually choose one
Choosing an ICP is not about finding the biggest market. It is about finding the segment where the next ninety days move fastest. Three questions do most of the work.
- Where do you win fastest? Look at your recent deals and ask which type closed with the least convincing. Short sales cycles are the market telling you where the pain is already felt.
- Where are the economics best? Deal size, margin, retention, expansion. A segment that closes easily but churns in three months is not your ICP, it is a distraction with good manners.
- Where can you prove it? Pick the segment where your existing work, references and results are most believable. Proof you already hold beats a market you would have to earn from zero.
Where those three overlap, that is your one. Write it down in a single sentence, with the role, the company type and the problem, and make it the first line of the plan.
Write the anti-ICP while you are at it
The list of who you will not chase is just as useful as the list of who you will. The deals that drag for months, haggle on price, and churn early usually share a profile. Name it. An anti-ICP gives you permission to say no quickly, and saying no quickly is where most of the focus actually comes from. Without it, the first tempting off-profile lead pulls the whole motion sideways again.
One ICP makes every downstream call easy
Here is the part people miss. Narrowness is not the sacrifice, it is the speed advantage. With one ICP, channel selection stops being a debate, because you go where those people already pay attention. List building becomes a filter instead of a fishing trip. The message writes itself in the buyer's own words. Ad targeting tightens, content topics become obvious, and every result is a clean read because you changed one variable at a time.
You are not married to it forever. You are choosing where to win first, then widening from a position of strength. If you cannot state your ICP in one sentence today, that is the first thing to fix, before another dollar goes into any channel. It is also the first thing we build in a GTM strategy engagement, so if the plan feels stuck, bring it to a call and we will pull it apart together.