Your Best Clients Are the Most Likely to Leave.

Churn Early Warning · Personal perspective

36% of MSPs have client retention rates below 50%. That means more than a third of the industry is replacing half its client base every year just to stay flat. All that sales effort, all that onboarding investment — not building toward anything. Just keeping the lights on.

Average annual MSP churn sits at approximately 12%, benchmarked across multiple industry sources. That number sounds manageable until you price in acquisition cost — which runs 5 to 25 times more than retaining an existing client, per research cited by the Harvard Business Review.

I spent 15 years selling enterprise security at RSA, Microsoft, Mandiant, and Palo Alto. In every one of those markets, retention was the silent variable that determined whether a business actually compounded or ran on a treadmill. The MSP market is no different. But what the latest data says about which clients are leaving is something most operators aren't prepared for.

MSPCFO recently published findings from an analysis of churn data across hundreds of MSPs. The results stopped me in my tracks — not because churn is a new topic, but because the specific patterns challenge everything operators assume about which clients are at risk.

High-efficiency clients — the accounts where the MSP runs lean, delivers reliably, and margins are strong — are churning at a 30% higher rate. The clients you run the tightest operation for are the ones walking out the door.

And clients without active project work in the past year churn at a 60% higher rate than clients engaged in four or more projects. Projects aren't just revenue — they're evidence of a future. A client with no active roadmap has already started mentally disconnecting.

There's a counterweight in the data too: MSPs with formal ticket budgets — operationally disciplined firms — show roughly 50% lower churn across the board. Structure on the operator side compounds. Structure on the client side does too.

The signals are there before the client makes a decision. Ticket volume drops. QBR attendance gets spotty. Recommendation uptake slows. No new projects get scoped. Engagement goes quiet. These aren't subtle — they're measurable patterns sitting inside the PSA, RMM, and billing data every MSP already collects.

The gap is that nobody's watching them systematically. Account managers are reactive. The QBR is quarterly. The client calls when they've already made a decision. A 5-point improvement in retention — from 90% to 95% — can lift profits by 25 to 95%, per HBR. That's not a rounding error. That's whether the business compounds or treads water.

The fix isn't more headcount. The data is already there. It just isn't being read in time.

This is exactly what we're building at Catalyst Shift with our Churn Early Warning module — M.02 of Catalyst OS. It runs on an MSP's existing PSA and RMM data. No new stack. No additional headcount. Just the visibility that makes retention proactive instead of reactive.

Weighted risk scoring across engagement data, ticket patterns, project activity, QBR history, and contract timing — surfaced as actionable flags before the client has made a decision. The clients most at risk are often the ones generating the best margins.