Bad segmentation used to be survivable. Now it kills your next GTM move.

Every GTM move starts with a choice about who. A new market, a new product, a push upmarket, a launch. Before anyone writes a message or plans a campaign, someone has decided which companies this is for.

Usually, that choice is made quickly. The segments already exist in the CRM. The ICP slide is from last year’s workshop. The next GTM move gets planned on top of them.

For a long time, that was good enough. If your segments were a bit too broad, the product made up for it. That’s changing, and it makes segmentation one of the most important decisions in any GTM move.

Segmentation used to be forgiving

When your product could do things your competitors’ couldn’t, you could afford a loose segment. You won the companies that needed those capabilities, and the rest of the segment was noise. The product did the sorting for you.

That’s no longer true. AI has made capabilities much faster to build and to copy. One of the places differentiation lives now is who you’re built for.

That’s a segmentation decision.

Why it matters more now

• Focus is your differentiation
  When products look alike, buyers choose the vendor that’s obviously built for them. You can only be the obvious choice for a group you’ve defined sharply. For everyone else, you’re one of several options.
• Buyers expect you to understand their world
  The workflow, the exceptions, the rules of their industry, what goes wrong after go-live. You can build that understanding for a segment. Not for everyone.
• Segments are fragmenting
  Companies that look alike on paper behave less alike. AI is a big reason: two companies with the same profile can be at completely different stages, with different problems and alternatives. But it’s not only AI. Regulation, the existing stack, the ability to build instead of buy and the growth stage can split a segment too.
• A GTM move amplifies the choice
  Your messaging, target account lists, content, sales effort and even your roadmap follow from it. Get the segment slightly wrong, and every one of those is aimed at a group that doesn’t behave as one.

Most segments are inherited, not tested

Most companies never chose their segments deliberately. They inherited them: from the CRM setup, a board deck, or the brackets a data provider happens to use. Nobody checked whether those groups actually buy differently.

For one client, companies with 200 to 1,000 employees were treated as a single segment. When we looked at the deal data, there was a sharp split after 400 employees. Below 400, the win rate was much higher. Above it, the win rate dropped off a cliff.

On paper, it was one segment. In the data, it was two markets: one they were winning, and one that took sales effort without much to show for it. Every campaign, every target account list and every forecast treated them as the same.

So the question isn’t “is this a sensible way to group companies?” It’s “does this grouping predict who we win?” That’s the first thing I look at in ICP and segmentation work.

What getting it right gives you

• You can predict where you’ll win
  Win rates by segment show where your product and proposition already work, and where they don’t. That’s a better basis for a GTM move than a big total number.
• Marketing and sales focus on the same companies
  Target account lists, campaigns and sales effort go to the segments you actually win, not spread across everyone who fits on paper.
• Your messaging gets sharper, and win rates go up
  Once you know which segments behave differently, you can speak to each one’s problem, urgency and alternatives. A message written for a segment lands better than one written for everyone in it.

Winnable market, not total market

GTM moves are often justified with a big number: 25,000 companies match our criteria.

But the number that decides whether the move works is smaller. How many of those companies have similar enough problems, urgency, expectations and alternatives that you can win them with the same product and the same proposition?

That’s your winnable market. Not who could buy in theory, but who you can credibly win right now.

A smaller winnable market isn’t bad news. It tells you where to put your effort first. And narrowing your market rarely shrinks your opportunity the way it feels like it will.

Segments you revisit, not set once

Segmentation often happens once: an ICP workshop, a slide, and it’s done for the year.

But buyers move. A company can go from experimenting with a new technology to depending on it within a few months, and its problems, expectations and alternatives change with it. A segment describes not just what a company is, but where it is right now.

So keep testing. Before every GTM move, and whenever win rates start to drift, ask whether the variables you segment on still predict who buys.

Before your next GTM move

1. Which segment is this move for?
   Not “mid-market” or “enterprise”. Specific enough that sales could build a list from it.
2. Does our segmentation predict who we win?
   Check your win rates by segment. If they vary a lot within one segment, it’s probably more than one.
3. Is it winnable now?
   Do these companies have the problem, the urgency and the expectations our product meets today?
4. Where are we the obvious choice?
   And where are we just one of the options?

The obvious choice for whom?

Your next GTM move will be judged on its results: pipeline, win rates, revenue. Much of that is decided before the first campaign goes out, by the choice of who it’s for.

The question isn’t who could buy. It’s who you’re the obvious choice for, right now.

Where I can helpMarket & Customer Intelligence