Coderhouse goes global

🌍

Read more

What Is an Ideal Customer Profile (ICP)? How GTM Engineers Turn It into Targeting Rules

AI Engineering

Build and ship real applications with LLMs, agents and automation.

8 weeks · Live online · View course →

Data Analytics

Go from zero to job-ready analyst with SQL, Python and BI tools.

10 weeks · Live online · View course →

Dan Patiño

AI Strategy & Innovation at Coderhouse

Artificial Intelligence

What Is an Ideal Customer Profile (ICP)? How GTM Engineers Turn It into Targeting Rules

What Is an Ideal Customer Profile (ICP)? How GTM Engineers Turn It into Targeting Rules

An ideal customer profile (ICP) is a description of the type of company that gets the most value from your product and brings the most value back to your business. It is not a single customer. It is a pattern: the industries, company sizes, tech stacks, business models and situations where your best accounts tend to come from.

For a GTM engineer, an ICP is more than a slide in a sales deck. It is a specification. Once the team agrees on who the ideal customer is, the GTM engineer turns that definition into data fields, filters, scores and automations that decide which accounts enter outbound sequences, which leads get routed to sales and which ones are deprioritized.

ICP vs buyer persona

The two terms are often mixed up. An ICP describes the account: the company. A buyer persona describes the people inside that company: their role, goals, pain points and how they evaluate solutions. In B2B, you usually need both. The ICP tells you which companies to target; personas tell you who to contact there and what to say. Salesforce's ICP guide and Clay's glossary entry both draw this same distinction.

What goes into an ICP

  • Firmographics: industry, employee count, revenue band, geography and company stage. See firmographics for the classic definition.

  • Technographics: the tools the company already uses, such as a specific CRM, data warehouse or ecommerce platform.

  • Business model signals: B2B vs B2C, self-serve vs sales-led, subscription vs transactional.

  • Situational triggers: recent funding, hiring for a specific role, a new executive, expansion into a new market.

  • Disqualifiers: conditions that make a company a poor fit no matter how large, such as regulatory limits or a missing integration.

How to build an ICP from data, not opinions

1. Start from closed-won and retained customers

Pull your best customers from the CRM, ideally those with high retention, fast sales cycles and healthy expansion. If you already measure customer value, techniques like RFM analysis help you separate the accounts that matter from the ones that merely signed.

2. Enrich the records

CRMs rarely have complete firmographic or technographic data. Enrichment tools fill in missing fields like industry, headcount and tech stack. Our guide on how GTM engineers use Clay for lead enrichment covers how teams combine several data providers.

3. Compare winners against everyone else

Look for attributes that appear far more often among your best accounts than in your total addressable market or in lost deals. A simple comparison table of win rate by segment is often enough to reveal the pattern.

4. Write the ICP as explicit rules

Translate the findings into criteria a system can evaluate, for example:

  • Industry in software, fintech or ecommerce.

  • Between 50 and 1,000 employees.

  • Uses a modern CRM and a cloud data warehouse.

  • Hiring for at least one revenue operations or sales role.

  • Not a government entity.

These values are illustrative; your own thresholds should come from your data.

5. Tier the accounts

Few accounts match every criterion. Tiering (for example Tier 1 perfect fit, Tier 2 strong fit, Tier 3 partial fit) lets you spend personalized effort where it counts and automate lighter touches elsewhere.

How GTM engineers operationalize the ICP

This is where the role differs from classic marketing work. A GTM engineer takes the ICP and wires it into the go-to-market stack:

  • CRM properties: create fields for ICP tier and fit score so every account carries its classification. Most CRMs support custom properties for exactly this.

  • Fit scoring: combine ICP fit with engagement signals into a lead score. See how GTM engineers build lead scoring models.

  • List building: filter prospect databases with the ICP rules to generate target account lists automatically.

  • Routing: send Tier 1 inbound leads straight to sales and Tier 3 leads into nurture.

  • AI classification: use an LLM to read a company's website and classify fuzzy criteria like "sells to developers" that structured data cannot capture, with outputs checked against labeled examples.

Keeping the ICP alive

An ICP is a hypothesis, and markets change. Review it on a regular cadence: compare win rates and retention by tier, check whether Tier 1 accounts really convert better and adjust the rules when they do not. Track the share of pipeline that comes from ICP accounts as a simple health metric for the whole GTM motion.

Common mistakes

  • Defining the ICP by aspiration. "Enterprise logos" is a goal, not a pattern in your data.

  • Too broad to filter. If almost every company qualifies, the ICP is not doing its job.

  • Never encoding it. An ICP that lives only in a document does not change who gets contacted.

  • Ignoring disqualifiers. Knowing who not to target saves as much effort as knowing who to target.

FAQ

How many ICPs should a company have?

Early-stage companies usually do best with one. Larger companies with distinct products or segments may maintain a few, each with its own rules and messaging.

Is the ICP the same as the total addressable market?

No. The total addressable market is everyone who could buy. The ICP is the subset most likely to buy, succeed and stay.

Can AI build an ICP for me?

AI can speed up enrichment, clustering and classification, but the ICP should still be validated against your real outcomes like win rate and retention.

Keep learning with Coderhouse

Turning a strategy into data, rules and automations is the heart of GTM engineering. The GTM Engineer Course is in early access, and you can join the waitlist today. To strengthen the analysis and AI skills behind targeting and scoring, explore the other two paths:

Dan Patiño

I'm Dan Patiño, head of AI Strategy & Innovation at Coderhouse. My day-to-day work involves merging the tactical management of e-commerce (CRO, Email Marketing and SEO) with the development of disruptive solutions. I specialize in building internal AI-powered apps to automate tasks and boost innovation within the team. I firmly believe that technology is strategy's best ally. To dive deeper into my professional journey, I'll be waiting for you on my LinkedIn profile.