TL;DR: Most ICPs are built using surface-level filters like industry, company size, or revenue. That helps identify companies that look similar, but not necessarily companies facing the problems your product solves best. The goal: identify and act on deals where your product creates disproportionate value.
An ICP definition sounds reasonable - the formula is: take your best customers, look for patterns, and try to turn those patterns into a repeatable definition. Industry, revenue range, employee count, geography, with maybe a few technology filters on top.
And that’s useful. Until the list grows to thousands of accounts that technically qualify, but actually the list becomes loose.
The outreach is going out to accounts, meetings are happening too - but a surprising number of “ideal customers” don’t really progress. While some take calls, many disappear after evaluation and others were never serious to begin with.
Leaders believe either sales didn’t stay on top of it, or the messaging needs work. Sometimes that’s true, but often the issue is elsewhere.
Problem 1: Why broad ICPs create false confidence
The blindspot is that often ICPs are built around similarity, not challenges. Even if a company resembles existing customers externally, that says very little about what is actually happening inside the business.
Let’s take an example: A commercial real estate company managing hundreds of properties absolutely fits the target profile for a facilities platform - they have scale, the operational complexity and the budget too. But one company may already have deeply embedded workflows, mature systems, and very little appetite for operational change. Another may be struggling with fragmented visibility across properties, rising tenant expectations, and growing pressure to modernize.
From the outside, both companies look nearly identical but they have very different maturity, and therefore need for change.
This is where broad ICPs create false confidence and waste rep time. Once a company matches the profile, it gets treated as equally relevant to every other account in that segment. But in practice, some accounts naturally fit your product’s strengths more than others.
This has less to do with firmographics and more to do with operational context.
These questions: what stage is the company in, what pressures exist internally, what goals are leadership driving toward, what is broken or becoming difficult to manage - reveal far more than employee count or revenue bracket.
The strongest GTM teams eventually think this way, even if they don’t formally call it jobs-to-be-done. They start asking: “Does this company resemble the situations where our product creates unusual value?”
Because the reality is that the same company can be an A-grade account for one vendor and a B-grade account for another, because each product aligns differently to the customer’s actual priorities and pain points.
Some products become valuable when operational complexity reaches a certain threshold. Others work best during periods of consolidation, expansion, or process breakdown. Some solve coordination problems, while others solve visibility problems.
Traditional ICP definitions miss those nuances, but those are exactly what determine whether a deal moves forward. Many account lists also become bloated over time - as the definition of fit keeps expanding. Soon, thousands of companies technically qualify, even though only a fraction are naturally aligned to what the product actually does best.
Even teams that solve the first problem: a sharper ICP definition - often hit the second one.
Problem 2: Why a good ICP definition isn't enough
A common problem is that GTM execution forgets the strategy. When the rubber meets the road, the ICP defined in slides stays there. Under sales and performance pressure, teams revert to intuition, prioritizing accounts based on familiarity.
This is where strategy needs to become a living system, one that translates into every execution decision, not just a definition that gets referenced once and then ignored. (We've been building something that makes this possible. More on that soon.) GTM teams get clearly graded accounts, and know where to spend their limited energy. No more false urgency for every account. This is where ICP scoring turns a static definition into a live execution tool.
Stronger ICP execution brings discipline by narrowing the focus, filtering for the cases where the product succeeds time and time again, where implementation friction is low and adoption happens naturally. These are accounts where customers are feeling the pressure the product is designed to relieve.
That kind of fit requires foundational work done deliberately, creating an advantage that creates better outcomes with less stress.
Does your ICP live in a Notion doc, or actually inform your team’s workflow?
Our next product update changes this.
FAQs
Q: Why do broad ICPs lead to poor pipeline conversion?
A: Broad ICPs are built around firmographic similarity: industry, size, revenue; rather than operational fit. Two companies can look identical externally but have completely different levels of urgency, maturity, and appetite for change. When every matching company is treated as equally relevant, teams waste cycles on accounts that were never genuinely aligned to what the product solves.
Q: What is the difference between ICP fit and problem fit?
A: ICP fit describes whether a company resembles your best customers on paper. Problem fit describes whether a company is currently experiencing the specific pressures your product is designed to relieve. A company can have strong ICP fit and weak problem fit - which is why many "ideal" accounts never progress past initial conversations.
Q: How do you move from ICP definition to ICP execution?
A: Most teams store their ICP in a document that is used during onboarding and then rarely consulted. Execution defaults to individual intuition. Closing this gap requires translating the ICP definition into a live ICP scoring system: one that grades accounts based on fit criteria and surfaces the highest-priority accounts continuously, not just when a rep is building a list.
