An ICP is a company-level definition, not a buyer-level one. It answers one question: of every company in the world, which ones should sales, marketing, and success spend their time on, and which should they politely pass on.
An ICP (ideal customer profile) is a description of the type of company that gets the most value from your product and gives you the most back in revenue, retention, and referrals. It combines firmographic traits (industry, size, geography), technographic signals (what tools they already use), behavioral patterns (how they buy), and outcomes (what they achieve). A strong ICP focuses go-to-market, raises win rates, and lifts lifetime value.
The five things to know before writing an ICP, and the one mistake that turns the exercise into a document nobody reads.
An ICP describes the type of organization you serve best: industry, size, revenue band, geography, tech stack, business model, and growth stage. It is the account you want to land, not the person who signs the contract. The buyer inside that account is a persona, and the two documents sit next to each other.
Firmographic traits (industry, headcount, revenue, geography) describe who the company is. Technographic signals (CRM, ERP, billing, support tools) describe what they already run. Behavioral patterns (how they research, how they buy, who they involve) describe how they decide. Outcome fit (the result they get after six months) is the final filter.
An ICP is the discipline of saying no. The teams that pick a narrow ICP outperform the teams that chase every lead, because every rep call, every ad dollar, and every onboarding hour is spent on accounts that will stay, pay, and renew. Chasing the whole market is the fastest way to lose to a competitor that chose one slice of it.
The ICP is not a workshop invention. It is a pattern you find in your closed-won data: which accounts had the highest LTV, the shortest sales cycle, the lowest churn, and the highest expansion. Score the fit factors across that cohort, test the pattern with sales for three months, and update when the data moves.
An ICP that lives in a Google Doc gets ignored by Friday. An ICP that lives as structured fields on the account record (industry, size, tech stack, segment, tier) drives routing, scoring, segmentation, and reporting every day. The CRM is where an ICP earns its keep. The doc is where it dies.
The two patterns that kill an ICP: writing it so wide every lead qualifies (which means none do), and writing it once and never revisiting. Markets move. Your product moves. The ICP has to move with them. A quarterly review tied to closed-won analysis is the minimum healthy cadence.
A strong ICP is not a sentence. It is a stack of filters that any new account can be measured against in under a minute. The four layers below are what separate an ICP that routes leads from an ICP that sits in a shared drive. Each layer answers a different question, and together they turn fit into something a workflow can score.
Industry or vertical, employee count, annual revenue, geography, legal structure, and growth stage. The baseline filters. If your product is built for 50 to 500 person US SaaS companies, that is the top of the funnel definition and the first thing any inbound form should capture.
The CRM, ERP, billing system, support tool, marketing platform, and data stack they run today. Technographic fit tells you whether the integration surface exists, whether the pain is real, and whether the switching cost is tolerable. It is one of the strongest signals of buying intent that nobody talks about.
Who they involve (champion, economic buyer, legal, security, procurement), how long the cycle usually runs, how they prefer to evaluate (demo, trial, POC), and the signals they send before they ever fill a form. Behavioral patterns are what separate the account that closes from the one that goes quiet after the first call.
The concrete result a good-fit customer gets: pipeline coverage up by X, cycle time down by Y, hours saved per rep per week, revenue attributed to the platform. If you cannot describe the outcome for a specific account, that account is probably not in the ICP, no matter how clean the firmographic match looks.
The average contract value, the typical sales cycle, the usual starting team size, and the expansion motion once they are live. An account that fits every other filter but closes at a third of your average ACV and churns at month nine is not an ICP account. It is a distraction wearing the right firmographic hat.
The complement of the ICP. Industries, sizes, business models, or tech stacks where you lose, under-deliver, or churn. Writing the anti-ICP is the hard half of the exercise. It is also where the real focus comes from, because declining the wrong-fit account is faster than untangling it twelve months later.
The most common mistake in go-to-market planning is treating the ICP and the buyer persona as the same thing. They are not. The ICP picks the companies you pursue. The persona picks the people you speak to inside those companies. Confusing the two produces marketing that targets the wrong accounts and sales motions that target the wrong buyers.
Industry, size, revenue, geography, tech stack, business model. Used to score inbound leads, route accounts, segment territories, and decide which outbound lists to build. The ICP tells the SDR team which 1,500 accounts to work and which 50,000 to skip.
Job title, role, seniority, pain, goal, source of truth, and preferred evaluation style. One ICP company usually has three to six personas you sell to (champion, economic buyer, user, security, procurement). The persona drives messaging, email copy, demo flow, and content strategy.
ICP picks the account. Persona picks the person inside it. Marketing campaigns target the persona, but only within accounts that match the ICP. Sales outreach targets the persona, but only when the account qualifies. Skip the ICP and your persona work waters down across every company on the planet.
Fit score uses ICP data (does this company look like our best customers?). Intent score uses persona and behavioral data (is the right buyer showing interest right now?). The two multiply. A high-fit account with a low-intent persona is a nurture play. A low-fit account with a high-intent persona is a polite pass.
The ICP fields live on the account record (industry, size, segment, tech stack, tier). The persona lives on the contact record (title, role category, buying role, pain tags). The reporting layer joins them so you can filter deals by ICP segment and persona at the same time without rebuilding the data every quarter.
The ICP is reviewed by RevOps and leadership quarterly, tied to closed-won analysis. Personas are reviewed by marketing and sales enablement, tied to messaging tests, call recordings, and buyer research. Both documents are living, but they move at different speeds for different reasons.
An ICP exercise fails when it starts with a workshop and ends with a doc. It succeeds when it starts with closed-won data and ends with CRM fields that score every new account. The sequence below is what the best RevOps teams run every quarter. It takes a week of focused work, not a six-month project.
Export the top 20 to 50 accounts by LTV, lowest churn, shortest sales cycle, and highest expansion. Four lists. The accounts that show up on three of the four are your target cohort. These are the companies the ICP needs to describe, before any opinion gets added.
For every account in the cohort, capture firmographic, technographic, behavioral, and outcome data. Count the patterns. Which industries show up most? Which tech stacks? What revenue band? Which deal shapes? The ICP is the shape that emerges, not the shape you wanted to see.
A one-page document with the four layers and the anti-ICP. Specific numbers, not ranges wide enough to swallow the market. "50 to 500 person US B2B SaaS running HubSpot or Salesforce, selling to mid-market, Series B through late growth." If a competing company can read it and disagree on who qualifies, it is too broad.
Route new leads through the ICP as a scoring layer. Measure win rate, cycle time, and ACV for ICP-fit accounts versus everything else. Pull the sales team into weekly reviews of the misfires: the ICP accounts that lost and the non-ICP accounts that won. The pattern in those misfires tunes the definition.
The ICP fields go on the account record. Scoring runs against them. Routing uses them. Dashboards segment by them. The ICP is now a system, not a slide. Quarterly reviews refresh it against the latest closed-won cohort, and the fields move with the pattern instead of chasing it.
Marketing targets ICP accounts in paid, outbound, and content. Sales works ICP accounts first and declines the rest politely. Customer success onboards ICP accounts to the outcome the ICP promises. Leadership forecasts off ICP coverage. Everything downstream gets sharper the moment the ICP stops being a document and starts being a filter.
An ICP that lives in a doc is a reference. An ICP that lives in the CRM is a system. The difference shows up in every pipeline review, every routing decision, and every quarterly board report. Below is the shape of a well-wired ICP in a modern CRM, and why each piece has to be structured data rather than freeform text.
Industry, size band, revenue band, geography, tech stack, business model, segment, and tier as structured fields on the account record. Picklists, not freeform text. The data has to be queryable, filterable, and reportable, which rules out the "ICP notes" text field that most teams start with and quickly abandon.
A fit score field on the account, calculated by rules that read the ICP fields and output a number or a tier. A good-fit account scores high, a non-ICP account scores low. The score updates automatically when the underlying fields change, so an account enriched from marketing becomes a tiered account for sales without manual work.
Lead and account routing rules that read the ICP tier and send high-fit accounts to the senior reps or the ABM pod, mid-fit to the standard team, and low-fit to a nurture or decline path. The ICP decides the SLA, the owner, and the follow-up pattern, not a round-robin that treats every account the same.
Audience segments that use the same ICP fields: paid ad audiences, email lists, retargeting pools, ABM target accounts. Marketing and sales use the same definition, so a campaign to ICP accounts gets routed to the same reps running the sales motion against those accounts. One data model, one go-to-market.
Dashboards that split pipeline, forecast, win rate, cycle time, and ACV by ICP tier. Leadership sees exactly where the quarter is being won: in-ICP coverage, in-ICP conversion, in-ICP expansion. Out-of-ICP activity shows up as a line to compare against, usually as a reason to recommit focus.
A quarterly review that pulls the latest closed-won cohort, re-scores the ICP against it, and updates the fields and the routing. The review is a workflow, not a workshop. The outputs are field changes, scoring rule updates, and a one-paragraph note to the GTM team about what moved and why.
Strkr stores ICP fields on the account record, scores every account against them, routes high-fit leads to the right reps, and reports pipeline by ICP tier. One tool, one data model, one definition of a good-fit customer, across sales, marketing, and success.
ICP stands for ideal customer profile. In go-to-market practice, it refers to a company-level definition of the type of organization that gets the most value from your product and returns the most value to your business in revenue, retention, and referrals. The term is used across sales, marketing, and customer success to align on which accounts the whole team should prioritize.
An ICP is a company-level definition (industry, size, revenue, geography, tech stack, business model). A buyer persona is a person-level definition inside that company (title, role, pain, goal, buying authority). One ICP company usually has three to six personas you sell to. The ICP picks the account; the persona picks the person inside the account. Both documents are needed, and they drive different parts of the go-to-market motion.
Start with your best customers, not a whiteboard. Pull the top 20 to 50 accounts by lifetime value, retention, cycle time, and expansion. Score the firmographic, technographic, behavioral, and outcome patterns across that cohort. Write a one-page definition with specific numbers, not broad ranges. Test the definition with sales for 90 days. Wire the fields into the CRM so every new account gets scored automatically. Refresh quarterly against new closed-won data.
Most companies land on one to three ICPs, segmented by product line, deal size, or motion (self-serve, mid-market, enterprise). More than three usually means the ICP exercise has drifted into segmentation. Fewer than one means focus is still forming. Early-stage companies often start with one tight ICP and expand only after they have won consistently in that first slice.
A usable ICP has four layers: firmographic traits (industry, size, revenue, geography), technographic signals (tools they already run), behavioral patterns (how they research and buy), and outcomes (the result a good-fit customer gets at month six). It also includes an anti-ICP that lists the companies you will decline. The document is one page. Anything longer is probably a strategy memo, not an ICP.
A quarterly review is the healthy cadence. Each quarter, pull the latest closed-won cohort, score the patterns, and compare them to the current ICP. Update fields, scoring rules, and routing when the data moves. Markets shift, products ship new capabilities, and the ICP has to shift with them. The teams that review the ICP once and never again are the teams that drift back into chasing every lead.
In the CRM, as structured fields on the account record, not in a Google Doc or a slide deck. Industry, size, segment, tier, and tech stack as picklists. A fit score field that reads those fields and tiers every account automatically. Routing, segmentation, and reporting all read the same fields. When the ICP is data, the whole go-to-market team uses it. When it is a doc, it is forgotten by Friday.
An ICP focuses the whole revenue motion. Marketing spends ad dollars on accounts that convert. Sales works the top of the list, not the bottom. Customer success onboards accounts that will stay. Leadership forecasts off a cohort that behaves predictably. The teams that pick a narrow ICP outperform the teams that chase every lead because every hour, dollar, and seat is spent on companies that will stay, pay, and expand.
No sales call, no migration consultant, no four-month implementation. Enter your card, get 14 days of the full Pro tier, cancel any time before day 14 with zero charge. Spin up a workspace, import your CSV, and have something useful before lunch.
