CRC = total retention cost in a period ÷ customers retained in that period
The arithmetic is easy. The split is not.
What counts as retention cost
- Customer success, the share aimed at keeping customers running rather
than growing them.
- Support, in full.
- Account management, the renewal-facing share.
- Retention programmes: onboarding, training, health monitoring, and the
tooling behind them.
- Renewal incentives and save discounts, which are a real cost and are
routinely omitted because they appear as reduced revenue rather than as spend.
The recurring judgement is the customer success split. A CSM who renews and expands is doing both jobs, and putting all of them in retention overstates CRC while putting none understates it. Pick a ratio, document it, and hold it across periods.
It overlaps gross margin on purpose
Most of these costs also sit in cost of revenue, so CRC and gross margin are two views of largely the same spend. That is fine. Margin asks whether delivery is efficient; CRC asks what a retained customer costs. Just do not present them as independent.
A worked example
A year in which retention functions cost $1,240,000 fully loaded, plus $110,000 of save discounts. 720 customers were retained.
CRC = (1,240,000 + 110,000) ÷ 720 = $1,875 per retained customer
Against an ARPA of $8,400, retention consumes about 22% of a customer's annual revenue. Whether that is acceptable depends on net dollar retention: high CRC buying strong expansion is an investment, while high CRC holding flat revenue is a subsidy.
What a rising CRC usually means
In rough order of likelihood:
- Targeting drifted. Customers acquired outside the ICP need more help to
stay, so acquisition volume is being paid for twice.
- Onboarding weakened, so support absorbs what enablement should have.
- A product gap is being covered by humans.
- Genuine investment in an expansion motion, which is the only benign
reason and the one everybody claims.
The first is the most common and the least often named, because it points upstream at Market rather than at the retention team.
The traps
- Omitting save discounts.
- Putting all of customer success in retention.
- Comparing to a benchmark without matching the function list.
- Dividing by all customers rather than customers retained.
- Reading it without ARPA or NDR, which give it scale and purpose.
Where this sits
CRC is Measurement in the Tenbound Pipeline Architecture Standard and it is the clearest financial expression of a Market failure: bad targeting shows up here as cost long before it shows up in churn.