Cornerstone

When growth plateaus after PMF, fix the system, not the symptom.

To diagnose a stall, confirm it against an equal earlier period, split growth into new pipeline, win rate and retention, and find which moved first. Plateaus look like channel problems; more often they are decision, evidence or ownership problems wearing channel clothes.

Get in touch Free · 20 minutes · No pitch Or, Take the Growth Bottleneck Scorecard → Compare services

Diagnose the stall in five checks, in this order

Each check rules something out, so the order matters. Confirm the stall is real, find which part of the number moved first, hear how buyers describe the problem, then test whether more effort helps or hurts. Only then decide what to change.

1. Same-length windows, one-off deals and seasonality removed
2. New pipeline, win rate or retention: which moved first
3. Ten recent sales calls: do buyers describe the problem your way
4. Would more spend make the weak part better, or louder
5. Retention falling? Check the product before the marketing

Force a constraint before you fund another quarter

The job is to name the lead constraint across positioning, acquisition, conversion, reporting, and ownership, then sequence work so the next expensive bet is obvious.

One lead layer
Evidence tied to revenue or qualified pipeline
A 30–90 day sequence
Falsifiable success signals

Then pick the engagement shape

Sometimes the answer is diagnosis only. Sometimes a sprint, a system build, or fractional leadership. The shape follows the constraint, not the other way around.

Growth Diagnosis when the path is fuzzy
Sprint when you need signal this quarter
System build when channels run in isolation
Fractional CMO when judgement cadence is missing

Questions

What founders usually ask here.

Short answers to the questions founders usually ask at this stage.

How do I diagnose why our startup's growth has stalled?

Start with data you already hold. Compare the last four to eight weeks with an equal earlier period and strip out one-off deals, seasonality and reporting changes, so you know the stall is real. Then split growth into new qualified pipeline, win rate and retention, and find the one that moved first: that is where the cause sits. Listen to ten recent sales calls and check whether buyers describe the problem the way your website does. Ask whether more spend would make the weak part better or louder; a capacity gap improves with more input, a positioning or conversion problem gets louder. If retention is what fell, growth work will not fix it and the honest next step is product and customer research. If you want an outside read, the Growth Diagnosis at We Scale Startups does this in one week from 90 days of CRM, GA4 and call data, for £2k–£4k.

Is a plateau always a marketing problem?

Often it is a system problem: positioning, offer, channel fit, conversion, reporting, or who owns scale/stop/fix. Product and PMF can still be the real block, diagnosis should say so honestly.

Do we need more experiments?

You need fewer untethered experiments and more tests anchored to a constraint map. See the experimentation page for how we structure learning once the bottleneck is named.

Notes that pair with this page.

Next step

Find the acquisition constraint before you add more activity.

Bring the current site, pipeline, campaigns, reporting, and team context. In 20 minutes we can usually tell which part of the system needs work first.

Get in touch Free · 20 minutes · No pitch Or, Take the Growth Bottleneck Scorecard →

Last updated: 6 August 2026