How to define an ICP that outbound can actually target
Most ICPs are positioning documents, not targeting instructions. Here is how to write one an SDR — or an agent — can turn into a list this afternoon.
Ask ten B2B teams for their ICP and you will get ten paragraphs like “mid-market companies who care about efficiency and want to modernise their go-to-market.” That is a positioning statement. Nobody can build a list from it.
A targeting ICP is a different document. It has to answer one question: given a company and a person, can I decide in under five seconds whether they belong?
The three layers
1. Firmographics — which companies
Observable, filterable facts. Industry, headcount band, region, funding stage, tech in use. Each one has to be checkable without a conversation.
Wrong: companies with a modern sales culture. Right: B2B SaaS, 50–500 employees, Series A–C, using HubSpot or Salesforce.
2. Persona — which human
Title alone is unreliable across companies. Anchor on what the person owns, then list the titles that usually own it.
Owns the number for outbound pipeline. Usually: Head of Sales, VP Sales, Head of Growth, RevOps lead. Not: CEO at 200+ headcount, individual AEs.
3. Triggers — why now
Firmographics say who could buy. Triggers say who might buy this month: hiring SDRs, new funding, a new sales leader, expanding into a region, or engaging with a competitor.
Triggers are what let you send twenty messages instead of two thousand.
Write the disqualifiers too
An ICP without exclusions is half a document. Ours reads better as a list of who to skip: under 20 employees (no budget), agencies reselling to their own clients (different sale entirely), pre-seed (buying nothing), companies whose last funding was over four years ago (usually flat).
Disqualifiers save more time than qualifiers, because they stop work before it starts.
Validate against closed-won, not intuition
Pull your last twenty closed-won deals and tabulate them against the ICP you just wrote. Two things usually fall out:
- A segment you win consistently but never deliberately targeted. Add it.
- A segment in your ICP with zero closed-won. Ask whether you have really tried it, or whether it is aspiration.
Then do the same for closed-lost and churn. A segment that buys quickly and churns in six months belongs in the disqualifiers, not the target list.
Keep it to one page
If the ICP does not fit on one page, it will not get used. Ours is a table:
| Layer | Include | Exclude |
|---|---|---|
| Company | B2B SaaS, 50–500, Series A–C | <20 staff, agencies, pre-seed |
| Buyer | Owns outbound pipeline | AEs, CEOs at 200+ |
| Trigger | Hiring SDRs, new funding, new sales lead | — |
That is enough for a person to build a list, and enough for an agent to score one.
Revisit it quarterly, not weekly
An ICP that changes every week is a guess. Set a review each quarter against the same closed-won analysis, and change it when the data says so — not when a single big deal comes in from outside it.
Related: how VSDR turns an ICP into booked meetings, and who this works for.
