Your GTM motion isn’t broken — your ICP is. Here’s the 4-variable diagnostic.
Your GTM motion isn’t underperforming because you picked the wrong channels. It’s underperforming because the company you built your ICP around in 2022 is not the company buying from you in 2026.
That gap — between the customer you designed your motion for and the customer you’re actually pitching — is the single most common source of pipeline waste in B2B. And it almost never shows up on a channel dashboard.
Why ICP drift happens (and why teams keep missing it)
The pattern is consistent across growth teams at 50–500 person companies: early traction comes from a narrow, well-understood buyer. The sales team knows exactly who picks up the phone, what they’re worried about, and how long it takes to close. Then the company scales. The TAM expansion argument wins the board meeting. The ICP definition gets stretched — “we can serve enterprise too” or “let’s add a new vertical” — and nobody formally revisits the fit criteria.
Win rate starts sliding. Average deal cycle stretches from 30 days to 90. The content that used to generate replies stops converting. The natural response is to blame the channel: the ads aren’t working, outbound open rates are down, SEO isn’t driving pipeline-quality traffic.
According to a 2023 Forrester B2B Revenue survey, companies that fail to systematically revisit ICP criteria waste an estimated 25–40% of marketing spend on segments that are structurally unlikely to close. The spend isn’t the problem — the targeting criteria feeding it is.
Drift is the mechanism. ICP refinement is the fix. But “refine your ICP” without a specific diagnostic is just advice. Here’s the diagnostic.
The 4-variable ICP diagnostic
You need four data pulls. If your CRM is reasonably clean, this takes a half-day. If it isn’t, the Growth Intelligence Engine can run the enrichment and segmentation work for you.
1. Fit score distribution across closed-won deals
Pull your last 50 closed-won deals and score each one against your stated ICP criteria: industry, company size, tech stack, geography, team structure, whatever your current profile says. Score each deal 1–5.
Most teams find a bimodal distribution: a cluster of 4s and 5s that closed fast and expanded, and a cluster of 2s and 3s that took three times as long and never grew past the initial contract. The 2s and 3s made it into the pipeline because the outbound or inbound motion didn’t filter them out early enough.
If more than 30% of your closed-won revenue came from accounts you’d score a 2 or 3 today, your ICP is wider than your sales motion can handle. That mismatch is where the waste lives.
2. Time-to-close by segment
This one is blunt: which segments actually close, and how long do they take?
Break your pipeline from the last 12 months into three or four segment buckets — by vertical, by company size, or by whatever dimension your team uses most. Calculate median time-to-close for each. The variation is almost always larger than people expect.
Intercom ran an internal analysis in 2019 that found their fastest-closing segment (mid-market SaaS, 50–200 employees) closed in 14 days on average. Their slowest segment (enterprise, 1000+) took 110 days. They were treating both as pipeline with equal urgency. Prioritizing by segment — not just deal size — changed how they staffed the sales function.
You don’t need Intercom’s scale to run the same logic. If one segment closes in 21 days and another takes 90, the second segment needs a different motion or a different position in the priority stack. Right now it’s probably competing for the same SDR hours and the same email sequences as the first.
3. Expansion rate by first-year cohort
Your real ICP isn’t who signs the contract. It’s who expands after month 6.
Pull every customer acquired in the last two years and calculate net revenue retention by cohort — grouped by the same segment dimensions you used in variable 2. Which cohorts are at 110%+ NRR? Which ones churned or contracted?
The expansion cohort is your ICP. Not the ideal customer you described in a workshop, but the customer your product actually serves well enough that they pay you more over time.
This is the most actionable variable of the four. It tells you which segment to weight in your outbound sequences, which case studies to lead with, and which verticals to cut from your paid targeting. The customers who expand are signal; everyone else is noise.
4. Churn signal by first-90-day behavior
Churn is almost always predictable at 90 days — the accounts that leave in month 8 or month 14 almost never looked healthy in month 2. They were just quiet.
Pull your churned accounts from the last 18 months. For each one, look at three indicators in the first 90 days post-close: product usage frequency, number of contacts engaged, and number of support tickets opened. Then compare those numbers to your expanded accounts.
The pattern that typically emerges: churned accounts had one power user and minimal team adoption. Expanded accounts had three or more active contacts in the tool by day 60. That’s not a customer success problem — it’s a qualification problem. The churned accounts were misfit at the point of sale.
Once you have that signal, you can build it into your sales qualification process: “Do they have three internal stakeholders who would use this?” becomes a hard gate, not a nice-to-have. That one change can drop first-year churn by 20–30 percentage points without touching the product.
What a refined ICP actually changes
ICP refinement isn’t a strategy document exercise. It changes three things immediately.
Channel allocation shifts. If your expansion cohort is 80% B2B services firms with 100–300 employees, and your current paid spend is weighted toward enterprise tech, you reallocate. Not partially — significantly. The data gives you the argument for the budget conversation.
Outreach changes in specificity. Generic outbound (“we help companies like yours grow”) collapses when the ICP is tight, because tight ICP means tight signal. You’re not writing sequences for “B2B companies” — you’re writing for “ops-led professional services firms adding headcount faster than their CRM can track.” That specificity changes reply rates more than subject line optimization ever will.
Content stops chasing traffic and starts attracting fit. If your ICP is a VP of Growth at a 150-person B2B services company, the article they read before booking a call with you is not a category overview. It’s a piece that describes their exact problem in language they use internally. You can only write that piece if you know who you’re writing for with that level of precision.
The four variables above give you that precision. The diagnostic takes a half-day. The channel strategy work that follows takes a week. The pipeline impact shows up in 60–90 days.
If you want to run this against your own data without doing it manually, start with the RF Studio Growth Diagnostic — it’s a structured audit that pulls the four variables and gives you a prioritized ICP refinement roadmap for your specific motion.