Most retention figures count customers. Net revenue retention counts money, and the difference matters because customers are not interchangeable.

Take the revenue from the customers you had twelve months ago. Look at what that same group is paying now. Add what they have upgraded or expanded; subtract what they have downgraded or canceled. Count nothing from anyone new. Divide by what they were paying a year ago.

Below 100% and the existing base shrinks, so new business has to refill the bucket before it fills it. Above 100% and the base grows on its own, which means new business is additive rather than remedial.

Why the threshold is the whole thing

A business at 90% has to win 10% of its revenue every year to stand still. That work is invisible in the headline growth figure and it is the first thing that stops working when the market gets harder or the sales effort is interrupted.

A business at 115% grows 15% a year with the sales team switched off. Not because it is better run, necessarily, but because its revenue is structurally different: the customers expand, and the expansion compounds. Run both for five years and the gap is a different business rather than 25 percentage points, which is compounding doing what it does.

This is why the figure attracts so much attention in acquisitions. It is close to a direct measurement of whether the revenue is an asset or a treadmill.

The two things it hides

It can conceal heavy customer loss. Lose 40% of your customers, but the 60% who stay are the large ones and they expand, and net revenue retention can still print above 100%. The figure is true and the business has a serious problem: it is failing most of the people it sells to and being rescued by concentration. Read it next to logo retention — the plain count of customers kept — and the picture is honest. Read it alone and it is not.

It rewards concentration. A business whose expansion comes from three large accounts has excellent net revenue retention and a structural fragility, because the same mechanism that produces the number produces the exposure. This is client concentration wearing a good metric as a disguise.

What actually moves it

Expansion built into the offer. Pricing that grows with the customer's usage, seats, or revenue means the account expands without a sale. Pricing that is flat per customer means every increase is a negotiation.

Reasons to stay that are not inertia. Accumulated data, integrations, trained staff, workflow built around the product — these are switching costs, and they are what turns a renewal from a decision into a default.

Not losing the large ones. Because the figure is revenue-weighted, one large cancellation can outweigh twenty small expansions. Which means attention should be revenue-weighted too — an unglamorous conclusion that most service businesses resist, because the small accounts are often the loudest.

The service business version

It is usually thought of as a subscription measure, and it applies perfectly well to anything with repeat revenue. For a consultancy: what did last year's clients spend this year? A firm whose clients come back larger has the same structural advantage as a software business at 115%, and most firms have never calculated it, which is why they discover in a bad quarter that their growth was entirely dependent on new logos.

It is also the cleanest available answer to whether the work is actually good. Churn tells you people left. Net revenue retention tells you whether the ones who stayed wanted more.