Skip to main content

​
What it is

logo_churn is the share of the customers you held at the start of a period who had left by the end of it. It is a count of customers, not an amount of revenue — a logo is one customer, whatever they were worth. It is a ratio rather than a percentage: 0.014 means 1.4% of the customers you started with had gone. It exists to be read against churned recurring revenue, and neither is complete alone. A period can lose many small customers and little revenue, or one large customer and a great deal of it. Those are different problems with different responses, and the only way to tell them apart is to read the count and the amount together. A quarter where this figure is high and revenue churn is low is a problem at the small end of your base; the reverse is a concentration problem.

​
How it is calculated

Measured.
A customer counts as churned on the same test used everywhere in Beacon: their recurring revenue reached zero. A customer who reduced what they pay has not churned, however far they reduced — that is contraction, and it does not appear here at all. Each customer counts once, whatever they were paying and however many subscriptions they held. The period can be a month, a quarter, a fiscal year, or the trailing twelve months. Short periods make this figure jumpy, especially with a small base: with 60 customers, one leaving is 1.7%.

​
Where it comes from

Your customer records for the count, and billing for the test that decides who churned. Billing is the required connection: without it there is no recurring revenue to reach zero, so no customer can be established as churned. Identity resolution decides what counts as one customer. Where two records for the same business are not resolved to each other, one of them ending reads as a churned customer even though the business is still yours.
No read interface or MCP tool serves this value yet — the table above publishes the shape ahead of the doors.
Why a customer count sits behind an opt-in. There is no money in the arithmetic, but what determines whether a customer churned is their recurring revenue reaching zero — so the figure carries the character of the revenue behind it rather than of the count on top of it. The opt-in is not keyed to money in any case: it follows how sensitive the record behind the value is, and every figure on the company and period object sits behind it on every door. Where the choice is made. This value is not open by default on any door. The choice happens when a reader grant that reaches it is issued or widened, when you subscribe to events, or when you turn the CRM property on. Whoever makes it is shown who becomes able to see the value, and their yes is recorded — holding a grant is no longer enough on its own.

​
How fresh it is

How often this refreshes before a period closes is not declared, and that is stated rather than filled in. A figure’s refresh cadence is the cadence of the lock that freezes it, read at the part of Beacon that owns the figure; this figure lives in Beacon’s shared core, which no single part owns, so there is nothing yet to read a cadence from. It will be filled in, not quietly. Once the period is closed the figure is fixed inside the locked record for that period.

​
Currency and rounding

No currency applies — it is a ratio of two counts. No rounding rule is set for this value. Beacon applies none of its own. A screen may show 1.4%; the published value is 0.014 at full precision, and comparisons should be made at the precision you received.

​
What changes it

Customers reaching zero recurring revenue during the period. The size of your base at the start of the period, which is the denominator — a growing base lowers this figure without anyone staying longer. A resolved identity, which can merge two records that would otherwise have counted as two customers. A customer reducing what they pay does not change it. A customer returning does not change it either; a return is reactivation, and it does not undo a churn already counted in an earlier period.

​
What it is not

  • Not churn_mrr. That is the revenue that left; this is how many customers left. Published together on purpose.
  • Not a retention rate. It is the share that left, not the share that stayed. One minus this figure is the customer-count retention rate, and Beacon does not publish that separately.
  • Not the inverse of gross_revenue_retention. That is revenue-weighted and includes contraction; this counts customers and ignores it.
  • Not churn_risk. That is forward-looking and per account; this is what already happened, across the company.
  • Not comparable across companies without care. A business with 40 enterprise customers and one with 4,000 small ones are not measuring the same thing with this figure.