Recurring revenue is not the same as predictable revenue
Recurring is treated as the gold standard. But recurring is not automatically predictable — and for an MSP building an AI practice, the difference decides what the business is worth.
Every MSP knows recurring revenue is the prize. It is why the managed services model exists, and it is the first number any acquirer asks for. But recurring is a description of how revenue arrives, not a promise that it will keep arriving. If customers churn, if every contract is bespoke, if renewals depend on somebody noticing in time — the revenue still swings.
Boards and buyers do not pay a premium for recurring. They pay it for predictable.
Two businesses, same label
Take two MSPs of the same size. Both describe their revenue as recurring. In the first, customers stay, contracts follow a consistent shape, and renewals are routine. In the second, churn runs high, every deal was negotiated separately, and each renewal is a scramble.
Both are recurring. Only one can be planned against.
Recurring is not predictable.
The revenue repeats. The number moves.
Churn is unmodelled. Contract terms vary customer by customer. Renewals happen because somebody chased them. The annual figure looks like an average of good quarters and bad ones, and nobody in the leadership team can say confidently what next year is.
Priced as services revenue.
The revenue repeats. So does the number.
Churn is understood and low. Contracts follow a standard shape with known terms. Renewal is a process with an owner and a date, not a rescue. The board can plan against it, and an acquirer can underwrite it.
Priced as a multiple.
Recurring tells you it repeats. Predictable tells you that you can count on it.
That distinction is the whole of it. A board can plan on predictable revenue. A buyer can underwrite it. Volatile recurring revenue offers neither — it carries the label without the confidence that should come with it, and the valuation reflects that whatever the headline ARR says.
New offers are where predictability gets lost.
This is the part that catches MSPs building an AI practice. Early AI work is almost always bespoke: every engagement scoped differently, priced differently, sold to a different sponsor for a different reason. That is reasonable in the first few deals, when you are still learning what the offer is.
It stops being reasonable at deal ten. Bespoke revenue does not compound — it just accumulates, and a portfolio of one-off AI engagements makes a business busier without making it more predictable or more valuable.
The fix is not to sell less AI. It is to convert what you have learned into an offer with a shape: a defined scope, a price, a delivery pattern, and a renewal that was designed in rather than hoped for. That is what turns AI work from project revenue into practice revenue.
Four things that steady the line.
None of them is a marketing exercise. All of them are commercial decisions the leadership team has to make.
Understand the churn you have
Not the headline rate. Which customers leave, at what point in the relationship, and for reasons you could have seen coming.
Standardise the contract shape
Terms, notice periods and uplift mechanics that are the same across the base. Variation is where predictability leaks out.
Make renewal a process
With an owner, a date, and a conversation that starts months before it expires rather than the week it does.
And the fourth, which is the hard one: package the new work. Take the AI engagements you have delivered, find the pattern, and turn it into a defined offer with a price. That is step two of the Five P System and it is the step most MSPs skip, because bespoke feels like responsiveness rather than the beginning of a valuation problem.
Four things worth remembering.
- Recurring revenue is not automatically predictable. The label describes how it arrives, not whether it keeps arriving.
- Churn, contract variation and renewal scrambles are what make recurring revenue swing.
- Boards and buyers pay a premium for predictability, not for the word recurring.
- AI work starts bespoke, which is fine at deal three and expensive at deal ten. Packaging it is what turns project revenue into practice revenue.
Is your recurring revenue actually predictable?
A Growth Review is a structured conversation with you and your leadership team: where your AI practice is constrained, what removing that constraint is worth, and whether your revenue can be planned against or only described. Ninety minutes, no cost, and an honest answer — including when the answer is that you do not need us.
Win the business case before you win the build.