Recurring-revenue businesses need two complementary views. Gross revenue retention measures how much starting recurring revenue remains after cancellations and downgrades, excluding expansion. Net revenue retention adds upgrades and expansion from the same starting cohort. Net retention can look healthy while gross retention deteriorates, so the pair should be read together with customer counts, reasons, segments, and revenue quality.
VISUAL LESSON
What you will learn
- 01Calculate GRR and NRR from a fixed recurring-revenue cohort.
- 02Build a waterfall for churn, contraction, and expansion.
- 03Segment retention drivers and assign operational experiments.
ILLUSTRATIVE MONTHLY WATERFALL
Expansion can offset loss without curing it
RETENTION WORKBOOK MAP
Build one cohort waterfall
List active customers and recurring revenue at the period start; exclude new customers acquired during the measurement period.
Record cancellation, downgrade or contraction, expansion, reactivation, plan migration, credit, and unresolved movement using agreed finance rules.
Compute retention for the same cohort, then break results by segment, product, tenure, location, acquisition source, and reason.
THE RETENTION WATERFALL
Starting revenue − churn − contraction + expansion
THE LEAD ATLAS METHOD
If retention analysis reveals a strong recurring-revenue segment worth acquiring, Lead Atlas Data can research business contacts matched to that segment’s locations, categories, market, and campaign rather than a broad prebuilt list.See how custom list research works ↗Freeze the starting cohort
Choose a month, quarter, or year and capture recurring revenue from customers active at the beginning. Exclude new customers added during the period from both GRR and NRR so acquisition does not masquerade as retention.
Define recurring revenue, customer, currency conversion, credits, pauses, failed payments, usage changes, plan migrations, and one-time fees. Reconcile the opening total with finance before calculating movement.
Calculate gross revenue retention
Subtract recurring revenue lost to cancellations and downgrades from starting recurring revenue, then divide by the starting amount. Do not add expansion. The result shows how much revenue the opening cohort retained before upsell.
Keep customer churn and revenue churn separate. Losing one large account can produce modest customer churn but severe revenue loss; losing many small accounts can produce the opposite pattern.
Calculate net revenue retention
Add expansion from the same starting customers after subtracting churn and contraction, then divide by starting recurring revenue. Do not include revenue from newly acquired customers.
NRR can exceed 100% when expansion is larger than losses. That is useful growth evidence, but it should not hide a declining GRR or a worsening experience among smaller customers.
Segment the waterfall
Break churn, contraction, and expansion by product, plan, tenure, acquisition source, customer size, location, onboarding path, implementation owner, voluntary versus involuntary loss, and reason code.
Validate reason codes with interviews, support records, product usage, payment events, and account notes. A generic reason such as budget is a starting clue, not a root cause.
Complete the retention action review
Assign the largest controllable loss driver to a named owner and one test: payment recovery, onboarding change, expectation setting, feature adoption, service intervention, plan redesign, or a better-fit acquisition rule. Protect the opening cohort definition while the test runs.
Deliverable: metric definitions, reconciled starting cohort, GRR and NRR formulas, revenue waterfall, segment table, top loss and expansion drivers, one retention experiment, guardrail, and next review date.
THE TAKEAWAY
Freeze a starting cohort, calculate gross losses before adding expansion, segment the drivers, and fix preventable churn even when net retention appears strong.OFFICIAL REFERENCES