ICP · Ideal Customer Profile
Ideal Customer Profile
What is an ICP (ideal customer profile)?
- Criteria for your best-fit 'ideal customer'
- Built from what your best customers have in common
- Focus resources where you can win
Once the ICP is written down, the first thing that changes isn’t what goes into the pipeline but the agreement on what gets taken out. The language of the sales meeting shifts from 'does this one look promising?' to 'how many of our criteria does this company meet?' The real change is that the basis for judging an opportunity moves from each rep’s gut feel to a shared company checklist.
Criteria are usually written in three layers: company attributes (industry, employee count, revenue, region); organizational and technical maturity (whether there is a dedicated team, which systems they already use); and the events happening at that company right now (triggers such as a new production line, regulatory response, or an executive change). Keep it to 5–7 criteria. Past ten, no company meets them all and the standard becomes meaningless. A buyer persona maps who reviews and who signs off inside that company, so it sits at a different layer. And while TAM and SAM are numbers that measure market size, an ICP is the filter that sets the order of priority within it.
The method is back-calculation, not hope. Rank your existing customers by days to close, renewal rate after one year, gross profit per deal, and number of support requests, then pull out the attributes shared by the top 20–30. A negative ICP comes out the same way. What the customers with early cancellations, late payments, and heavy customization demands have in common becomes your exclusion criteria. If your product lineup or pricing changes, recalculate at least once every six months so the criteria don't drift from reality.
The most common mistake in practice is writing the ICP as 'the big companies we want to sell to'. Put large conglomerate affiliates in your ICP and sales gets stuck in six-month review cycles and loses an entire quarter, while mid-sized companies that would sign in two months get pushed aside. Conversely, if you build an ICP but pay incentives only on total revenue, reps go back to chasing any deal that comes along. An ICP is worth something only when it works as a rejection rule, not a document. Set a firm cap in advance on the time you spend on poor-fit customers, such as out-of-criteria inquiries get a one-page quote instead of a full proposal.
John KimAuthor of B2B Marketing Design and CEO of Nanum Consulting
Jaehyuk ParkHands-on instructor in B2B technical sales and AI-driven sales strategyIt has no single originator. As ABM and data-driven B2B sales spread, the need to focus on the 'best-fit client' rather than 'anyone' made it a standard concept.
Consider a hypothetical case: Company A, a B2B training and consulting firm with 90 employees and KRW 13 billion (≈ US$9,726,200) in annual revenue. Six salespeople handled nearly 400 consultations in a year, yet the win rate stayed at 11%. Each proposal took an average of 3 days to write, and half of them went to companies that had not even confirmed a training budget.
- The VP of sales exported all 180 contracts from the past two years out of the CRM, sorted them by renewal status, gross profit per deal, and post-project satisfaction, and set aside the top 30 companies.
- Lined up side by side, those 30 companies revealed what they had in common: manufacturers and parts suppliers with 300–1,000 employees, at least one dedicated HRD staff member, an annual training budget of KRW 100 million (≈ US$74,800) or more, and in year two of a smart-factory push.
- Conversely, when they dug into the 12 early cancellations, 9 were companies with fewer than 100 employees or where the training manager doubles as general affairs. The director laid down the rule: 'if both of these apply, don’t write the proposal.'
- The marketing lead added employee count and a dedicated-HRD-team field to the inbound inquiry form, and rebuilt the scoring rules so that leads meeting both conditions get 30 points up front.
- Six months later, consultations fell from 400 to 230, but the win rate rose from 11% to 24%, the average contract value grew from KRW 18 million (≈ US$13,500) to KRW 31 (≈ US$1) million, and the sales cycle shrank from 74 days to 52.