Gross Revenue Retention
The percentage of contracted recurring revenue you keep from your existing customer base over a period, counting only churn and downgrades — never expansion.
◆ Percentage
Formula
Built from
What it measures
GRR isolates the contracted revenue you keep from the customers you already had at period start. It takes only the base you opened with and asks: what fraction survived churn and downgrades? It deliberately ignores every form of expansion — upsells, cross-sells, seat adds, price increases — so it can never exceed 100%. The gap between GRR and Net Revenue Retention is exactly how much your expansion engine added back on top of a shrinking base.
Why it matters
You track GRR because it answers one brutal question: are we keeping what we already have? A company with 100% ARR growth but weak GRR is running a leaky bucket — new bookings are masking that the base itself is eroding. Because GRR is dollar-weighted, it tells the truth that logo churn hides: losing one $50K customer hurts GRR far more than losing ten $1K customers. Boards read GRR as a leading indicator of churn risk and product stickiness, because it strips out the expansion revenue that NRR uses to flatter a soft retention story.
How to read it
Read GRR as the share of your opening contracted revenue pool that survived the period. 90% means you opened with $100K CARR and lost $10K to churn and downgrades, keeping $90K — and the most that's possible is 100%, since expansion is excluded by design. You want GRR high and stable or rising month over month. The single most useful read is the gap to NRR: if GRR is 90% and NRR is 110%, that 20-point spread is your net expansion, and it's the only thing keeping a leaky base afloat. Always compare to the prior period and your plan, not to one snapshot.
What good looks like
Good
High GRR with most of the opening base retained period over period, and the rate holding steady or improving — a sign of a sticky product and durable customer fit.
Watch
A meaningful slice of opening CARR leaking each period, or GRR drifting down month over month even if the absolute level still looks acceptable — investigate downsell and churn drivers.
Bad
A large share of the opening base lost each period, leaving the company dependent on heavy new bookings just to stand still — the base is unsustainable without product or pricing changes.
Watch-outs
- Letting new and expansion revenue sneak in. GRR counts only losses from the opening base — if a single dollar of upsell or new-logo CARR touches the numerator, the metric breaks and can read above 100%, which is mathematically impossible for true GRR.
- Conflating GRR with NRR. A sales leader celebrating a high NRR can miss that GRR is sliding underneath it; the base is eroding while expansion velocity hides the leak. Always track and report both, separately.
- Reading logo retention instead of dollar weight. 99% logo retention with weak GRR means your largest accounts are the ones leaving. Segment GRR by customer size and cohort or you'll miss the whales walking out.
- Measuring on the wrong base. Run GRR off your defined opening cohort's CARR, not a moving live-ARR figure that already nets in mid-period changes — mixing bases makes the period-over-period trend meaningless.
Worked example
Hypothetical
You open June with $1M in CARR. Two customers fully churn — one worth $100K, one worth $50K — for $150K of Churned CARR. A third downgrades from $200K to $150K, adding $50K of Downsell CARR. Your surviving base is $1M − $150K − $50K = $800K, so June GRR is $800K ÷ $1M = 80%. Two other customers expanded by $80K that month, but that's invisible to GRR — it only lifts NRR, to ($800K + $80K) ÷ $1M = 88%.
Variants & windows
The same metric re-expressed by a mechanical transform — a trailing window, a growth rate, a per-unit scaling, or a book/segment cut. Each is computed from Gross Revenue Retention above.
- GRC (MOM) Alternate cut of the parent metric
- GRC (T3M) Trailing 3-month
- GRC (TTM) Trailing 12-month