Revenue Churn vs. Customer Churn: Why Losing a Big One Isn't Losing a Small One

A founder once told me with relief that the company's churn had dropped from eight to five percent in the quarter. I asked to see the revenue lost over the same period. It had gone up. He was measuring heads and the cashier was bleeding from a place that his metric couldn't see. It lost fewer customers and more money, because those who left were the big ones.

This mismatch is more common than it seems, and it arises from treating two different numbers as if they were the same. Customer churn and revenue churn measure losses, but they measure different things. One counts how many relationships have broken. The other counts how much money went out the door. When your ticket is uniform, they go together. When you have clients of very different sizes, they can point in opposite directions, and then choosing the wrong one to look at is costly.

What each number is really counting

Customer churn, or logical churn, is a count of accounts. You take how many customers you had at the beginning of the period and see how many canceled. The result is a percentage of closed relationships. In this account, every customer is worth exactly one. The one paying for the lowest plan and the one paying for the largest contract add up to the same amount.

Revenue churn, or financial churn, changes the unit. Instead of counting bills, it adds up the recurring revenue that has left the base due to cancellations and plan reductions. Here each customer weighs the size of the check he wrote. Losing a big contract moves this number a lot. Losing a small account barely scratches the surface.

The consequence of this change of unit is everything. Customer churn responds “how many relationships am I losing.” Revenue churn answers “how much of my revenue am I losing.” These are different questions, and the health of the business depends much more on the second than the first, because it is the revenue that pays the bill, not the number of logos on the website.

It is worth fixing the connection with the base. Revenue churn is always a percentage of what? From your recurring revenue, the MRR. That's why this number only makes sense when you read your recurring revenue unambiguously first, a subject I explained in the text about MRR and monthly recurring revenue. Without a clean revenue base, any financial churn becomes a kick.

Why losing a big one hurts differently

The intuition that every customer is worth the same is the original error. In almost all SaaS, the distribution of revenue per customer is skewed. A minority of large accounts account for a disproportionate share of revenue, and a majority of small accounts account for little. When this is true, counting heads lies about the seriousness of what is happening.

Think about two losses in the same month. At first, ten small customers cancel. In the second, a single large customer leaves. According to the customer churn metric, the first situation is ten times worse. According to the revenue metric, it could be the opposite, if that single contract was worth more than the ten small ones combined. The same event, read by two rulers, produces two opposing diagnoses.

It changes where you put attention. If your revenue is concentrated in a few large contracts, your real risk is not the stampede of small ones, it is the silent exit of a large one. A client that represents a large portion of the base deserves treatment that a small client does not justify: close monitoring, direct relationship, anticipation of problems. Not because of favoritism, because of risk management. The concentration that got you up quickly is the same one that brings you down quickly if you don't look at it.

There is another side, equally important. An operation that loses many small customers but holds on to large ones can have healthy financial churn and scary customer churn. This isn't necessarily a revenue problem, but it's a signal about the product and who you're attracting. It may indicate that the top of the funnel brings people without fit, who activate poorly and leave early. The revenue number is good, but the customer number is warning you of something that will charge the bill down the road.

Gross revenue churn: the hole before any consolation

When talking about revenue churn, the most honest metric to start with is gross revenue churn. It only measures what came out: cancellations and plan reductions, without mixing it with anything positive. It's the raw photo of the leak.

The reason for insisting on gross is that there is a net version, net revenue churn, which removes from this hole what the base gained from expansions and upsells from those who stayed. The liquid version is powerful and deserves its own text, which I dedicated to Net Revenue Retention. But it has a dangerous side effect when used too soon: expansion can mask the leak.

Imagine a base where large customers are growing well, buying more seats, upgrading plans. This growth can offset, in aggregate, a relevant loss of revenue due to cancellation. The net number looks beautiful. And, hidden behind it, there is a high gross churn that is being financed by the expansion of a few. The day expansion slows, the hole that has always been there appears at once.

That's why I look at gross revenue churn first, always. It tells me how big the leak really is, regardless of how much the expansion is covering. Then I look at the liquid to understand whether the base, as a whole, is gaining or losing. The two readings together tell the complete story. Only the liquid tells a reassuring story that may be false.

How to use both metrics without making mistakes

The rule I follow is simple: never look at one alone. Customer churn and revenue churn are two lenses on the same phenomenon, and each sees what the other doesn't.

When the two rise together, the diagnosis is straightforward: you are losing relationships and money at the same rate, probably a transversal value or fit problem. When customer churn rises and revenue remains stable, you are losing the small ones and holding on to the big ones, which points to an acquisition quality problem at the bottom end. When revenue churn rises and customer churn remains stable, the most serious alarm sounds: few but large ones are leaving, and concentration is taking its toll.

Each of these scenarios calls for a different action, on a different team. Treating everything as "churn has increased" and throwing the same medicine at it is wasting effort in the wrong place. Segmentation is what turns a reporting number into a decision.

There is an additional layer worth monitoring: churn by customer size range. Looking at the loss separately between small, medium and large accounts reveals patterns that the aggregate number hides. Maybe the small ones cancel because of price and the big ones because of a lack of a specific feature. They are different causes, with different solutions, which only appear when you break the number down by segment instead of looking at the average.

What changes in the mind of those who decide

The question I ask those who report churn is always the same: is this number of customers or revenue? Anyone who doesn't know how to respond right away is looking at a single number and thinking they've looked at retention. Those who know will answer with both and with the breakdown by size, because they understand that the average hides more than it reveals.

The maturity of a SaaS operation is measured, in part, by whether this distinction is alive in meetings. Teams that only talk about "churn" in the singular have not yet internalized that losing a big one and losing a small one are events of completely different severity. Those who lead well force this separation until it becomes a reflex, because it is what protects the company from the false comfort of a number that has improved while the cash flow has worsened.

If you take one thing away from here, let it be this: count heads to understand your product and your funnel, count revenue to understand your survival. Both matter, but only one pays the bill at the end of the month.