Gross Revenue Retention: Formula & Benchmarks (2026)

Gross revenue retention tells you how much recurring revenue your existing customers keep paying before upgrades make the result look better. If you run a SaaS business, it is the cleanest way to see whether churn and downgrades are quietly eating your base.

The calculation is simple. Getting the cohort, revenue movements, and reporting period right is where teams make mistakes. This guide gives you the formula, a worked example, practical 2026 benchmarks, and a repeatable way to monitor GRR from your own database.

Gross revenue retention: the quick answer

Gross revenue retention, or GRR, is the percentage of recurring revenue retained from customers who existed at the start of a period. It subtracts revenue lost through cancellations and downgrades, excludes new customers, and gives no credit for expansion revenue.

GRR = (Starting recurring revenue - churned recurring revenue - contraction recurring revenue) / Starting recurring revenue × 100

Suppose you start a month with $100,000 in MRR. Cancellations remove $4,000 and downgrades remove another $3,000. Your GRR is ($100,000 - $4,000 - $3,000) / $100,000 × 100 = 93%. Any upsells during the month are deliberately ignored.

Why GRR catches problems that growth can hide

New sales can make total MRR rise while existing customers are leaving. Expansion can do the same inside your installed base. GRR strips both effects out, so a falling number points directly at retention quality.

That makes GRR useful for founders, finance teams, customer success leaders, and product managers. A customer success team can segment it by account tier or onboarding path. Product can compare it by feature adoption. Finance can use it to test how dependable recurring revenue really is.

GRR also has a natural ceiling of 100%. You cannot retain more starting revenue than you had without counting expansion, and expansion does not belong in this metric. A result above 100% means the calculation has mixed GRR with net revenue retention.

How to calculate gross revenue retention correctly

1. Freeze the starting customer cohort

List every paying customer active at the first instant of the measurement period. Only these accounts belong in the calculation. A customer acquired halfway through the month is new business, not retained revenue, so exclude that account from both numerator and denominator.

2. Record starting recurring revenue

Use MRR for a monthly calculation or ARR for an annual calculation, but do not mix them. Normalize annual, quarterly, and monthly contracts into the same recurring-revenue unit. Exclude setup fees, professional services, usage spikes, taxes, and other nonrecurring charges.

3. Subtract cancellations and downgrades

Churned recurring revenue is the starting MRR or ARR lost when a cohort customer cancels. Contraction is the decrease when that customer moves to a cheaper plan, removes seats, reduces committed usage, or receives a recurring discount. Count only losses that took effect during the period.

4. Ignore upgrades and new customers

An existing customer may upgrade from $500 to $800 per month. GRR still gives that account a maximum retained value of $500 because $500 was in the starting cohort. The additional $300 belongs in expansion MRR and affects NRR, not GRR.

A worked SaaS GRR example

A SaaS company begins Q2 with 200 customers and $120,000 in MRR. During the quarter, five customers representing $6,000 in MRR cancel. Ten others downgrade, reducing MRR by $3,600. Existing customers also add $11,000 in expansion MRR, and new customers add $20,000.

GRR = ($120,000 - $6,000 - $3,600) / $120,000 × 100 = 92%.

The $11,000 expansion and $20,000 new MRR do not enter the formula. Total MRR can still grow strongly, but the company lost 8% of its starting revenue base. That is the retention problem GRR makes visible.

Gross revenue retention benchmarks for 2026

Stripe's May 2026 retention guide reports that the median GRR across SaaS companies in 2025 was 91%. It describes 90% or higher as healthy for most SaaS businesses, 95% or higher as excellent, and 70% to 85% as a range that deserves attention.

Use those figures as orientation, not a universal target. Contract size, buyer type, billing cadence, and company maturity change what good looks like. ChartMogul's SaaS Retention Report found that top-quartile companies with average revenue per account above $500 per month exceeded 90% GRR, while top-quartile businesses below $50 per month were around 60% to 70%.

For a self-serve product with low monthly prices, 90% annual GRR may be exceptional. For an enterprise product with annual contracts and high switching costs, the same result may signal avoidable losses. Compare your company with peers that share your customer segment, average contract value, and measurement window.

Your own trend is more actionable than a broad benchmark. Track GRR every month, inspect rolling three- and twelve-month views, and segment by plan, acquisition channel, industry, account age, and customer success owner. A stable company-wide rate can hide one deteriorating segment.

GRR vs NRR: use both, but do not confuse them

Gross revenue retention excludes expansion. Net revenue retention includes it. If the example company adds $11,000 of expansion MRR, its NRR is ($120,000 - $6,000 - $3,600 + $11,000) / $120,000 × 100 = 101.2%, while GRR remains 92%.

NRR shows whether the existing base can grow after losses. GRR shows the losses before upsells offset them. A high NRR paired with weak GRR means expansion is carrying the business while other customers churn or contract. That may work temporarily, but it narrows the base available for future expansion.

How to monitor GRR from your database without SQL

A spreadsheet works for a one-off calculation. It becomes fragile once you need monthly cohorts, plan segments, and alerts. Your billing or application database already contains most of the required fields: customer ID, subscription status, recurring amount, plan, effective date, cancellation date, and account segment.

With AI for Database, you can connect PostgreSQL, MySQL, Supabase, BigQuery, MongoDB, or another supported database and ask: “For customers active on July 1, calculate July GRR. Exclude new customers and expansion. Show churn and contraction MRR by plan.” You get the answer without writing SQL or waiting for an analyst.

Save the result as a self-refreshing dashboard with GRR, churn MRR, contraction MRR, NRR, and cohort size. Then add an action workflow that emails or posts to Slack when monthly GRR falls below your target, a segment drops by more than two percentage points, or a high-value account downgrades.

That closes the loop between reporting and action. Instead of finding a retention problem in a quarterly board deck, your customer success team sees the movement when it happens and can investigate the affected accounts.

Five GRR mistakes that distort the result

Including new customers: this inflates retained revenue with sales that were never in the starting cohort.

Counting expansion: GRR cannot exceed 100%. Put upgrades and extra seats in NRR instead.

Mixing logo churn with revenue churn: losing 5% of customers does not mean losing 5% of revenue. Calculate GRR from recurring revenue, not account count.

Ignoring downgrades: a customer can stay active while cutting spend sharply. GRR must include contraction as well as full cancellations.

Comparing mismatched periods: monthly GRR and annual GRR are not interchangeable. Label the window clearly and compare like with like.

Questions SaaS teams ask about gross revenue retention

What is a good gross revenue retention rate for SaaS?

A practical general target is 90% or higher annually, while 95% or more is excellent for many SaaS companies. Your appropriate benchmark depends heavily on contract value, customer segment, and billing model.

Can gross revenue retention be over 100%?

No. GRR excludes expansion revenue, so 100% is its ceiling. If your result is higher, upgrades, new customers, or another revenue source has entered the calculation.

Should I calculate GRR monthly or annually?

Track monthly GRR for operational changes and rolling twelve-month GRR for long-term retention. Annual-contract businesses should also use renewal cohorts so accounts are measured when they are actually eligible to renew.

How can a non-technical team calculate GRR from live data?

Connect the billing or subscription database to a natural-language analytics tool, define the starting cohort and exclusions in plain English, and save the result as a recurring dashboard. Add alerts for threshold breaches so the metric prompts action.

Turn GRR into a retention operating metric

Calculate one reliable baseline, then segment it until the loss has an owner. If GRR drops, identify whether cancellations or downgrades caused it, which accounts moved, and what happened before the change. A metric without that path to action is just decoration.

AI for Database lets your team ask those follow-up questions against live database data, keep the result current on a dashboard, and trigger a workflow when retention slips. Connect your database at aifordatabase.com and build your first GRR view without writing SQL.

Sources

Stripe, “Net revenue retention vs. gross revenue retention,” updated May 28, 2026: https://stripe.com/resources/more/net-revenue-retention-vs-gross-revenue-retention

ChartMogul, “SaaS Retention Report”: https://chartmogul.com/reports/saas-retention-report/