SaaS Retention Benchmarks: How to Use Them

A benchmark can tell you that retention is weak. It can't tell you why. That takes the right metric, a fair peer group, and clean billing data.

Current research puts median NRR near 101%, median GRR near 84%, monthly churn near 3.5%, and annual churn near 35%. Use those figures as starting points, then compare them with your own stage, ACV, and customer mix.

Table of Contents

• Step 1: Choose the Retention Metric That Matches Your Question
• Step 2: Compare Your Numbers With the Right SaaS Retention Benchmarks
• Step 3: Calculate Your Retention Metrics From Stripe or Paddle Data
• Step 4: Segment Retention Before You Decide Whether It Is Good or Bad
• Step 5: Turn a Benchmark Gap Into a Retention Improvement Plan
• FAQ: SaaS Retention Benchmarks
• Conclusion

Step 1: Choose the Retention Metric That Matches Your Question

SaaS retention benchmarks only help when the metric matches the question you're asking. Start by naming the business problem before you open a dashboard.

Ask, “Are we losing customers?” Then track logo churn. Ask, “Are we losing revenue?” Track revenue churn and GRR. Ask, “Is our existing base growing?” Track NRR.

NRR includes churn, downgrades, and expansion from current customers. It excludes new customers. The formula is:

NRR = (Starting MRR + Expansion MRR - Contraction MRR - Churned MRR) ÷ Starting MRR × 100

GRR removes expansion from that equation. It shows how much starting revenue remains after cancellations and downgrades. GRR can't exceed 100%, which makes it a useful check against an NRR result that looks healthy only because of upsells.

For a clear explanation of the difference, read what net revenue retention means. The key point is simple: NRR shows growth inside the base, while GRR shows revenue loss.

Pick a reporting period that matches your billing model. Monthly plans need a monthly view. Annual contracts need renewal and annual retention views. You can still review monthly movements, but don't compare a monthly churn rate with an annual renewal rate.

• Logo churn: customers lost ÷ customers at the start.
• Revenue churn: recurring revenue lost ÷ starting recurring revenue.
• GRR: starting revenue after churn and contraction.
• NRR: starting revenue after losses and expansion.

Chartsy can show these metrics from Stripe and Paddle data, which helps small teams avoid rebuilding each measure in a spreadsheet. If your question is about customer value, add ARPU and LTV. If it is about acquisition quality, connect retention to the source that brought each customer.

Keep the metric list small at first. Five well-defined metrics beat a dashboard full of numbers nobody trusts.

Step 2: Compare Your Numbers With the Right SaaS Retention Benchmarks

SaaS retention benchmarks are comparison points, not pass-or-fail grades. Your first job is to match your number with companies that sell in a similar way.

The current research sample points to these broad figures:

Metric	Reported benchmark	How to use it
Net revenue retention	Median near 101%	Check whether the existing base grows or shrinks
Gross revenue retention	Median near 84%	Check revenue loss before expansion
Monthly churn	Median near 3.5%	Watch short-term customer loss
Annual churn	Median near 35%	Review the full-year effect of monthly churn
Annual customer retention	25th to 75th percentile: 88% to 90%	Use only with a similar contract cycle

These figures don't form one perfect dataset. They come from different studies with different samples and definitions. One source reports an unusual NRR spread, with the lower quartile above 110%. Treat that result as a signal to inspect the method, not as proof that every SaaS company should hit 110%.

Stage changes the comparison. CRV reports monthly logo churn of roughly 3% to 5% for companies below $1 million ARR. For companies between $1 million and $10 million ARR, annual churn commonly falls in the 10% to 15% range. Those are different lenses, so don't place them in one scorecard.

Customer size matters too. SMB customers often cancel more easily. CRV reports SMB GRR in the 70% to 82% range, while enterprise accounts can show much stronger revenue retention. An SMB product with 97% NRR may be doing well even if an enterprise benchmark says 110% is the target.

Use the SaaS churn rate benchmarks reference to record the metric definition, period, segment, and source beside every comparison. That small habit stops benchmark drift during board reviews. If NRR specifically is the number you're trying to interpret, what counts as a good NRR breaks that one metric down by stage and pricing model in more depth than this multi-metric view can.

Make a peer card for your business with four fields:

• ARR band
• Average contract value
• Monthly or annual billing
• Self-serve, sales-led, or mixed motion

Then compare your result with three baselines: the current benchmark, your last quarter, and your best prior cohort. A number below the market median may still show strong improvement. A number above it may hide a worsening trend.

For formal reporting, keep the source date and definition. Benchmark data is sparse. In the sample used for this article, only eight of seventeen items included a median, and fewer than half included both percentile values.

Step 3: Calculate Your Retention Metrics From Stripe or Paddle Data

To use retention benchmarks, first build one clean starting cohort. Your calculation is only as good as the accounts included in it.

Export the subscriptions that were active at the start of the period. Exclude new customers from that period. Then map each account's recurring revenue movement into four buckets:

1. Starting MRR from the existing customer base.
2. Expansion MRR from upgrades or added seats.
3. Contraction MRR from downgrades.
4. Churned MRR from full cancellations.

Use this equation:

NRR = (Starting MRR + Expansion MRR - Contraction MRR - Churned MRR) ÷ Starting MRR × 100 — see the net revenue retention formula for the standard calculation.

Imagine your starting MRR is $100,000. Existing customers add $8,000, downgrade by $2,000, and cancel $5,000. NRR is 101%. New customer revenue stays out of the calculation.

Paddle's NRR explanation also stresses this point: NRR measures recurring revenue from existing customers, not total new business. For a worked calculation process, review how to calculate net revenue retention.

Now calculate GRR by removing expansion:

GRR = (Starting MRR - Contraction MRR - Churned MRR) ÷ Starting MRR × 100

In the example, GRR is lower than NRR. That difference shows expansion is covering part of the lost revenue. That may be fine, but it deserves a closer look.

Check your work against billing records. Review cancellations, refunds, pauses, coupons, plan changes, and failed payments. Decide how each one is treated before you compare this month with last month.

Chartsy connects with Stripe and Paddle so founders can ask questions about MRR, churn, upgrades, downgrades, ARPU, and LTV without hand-sorting every transaction. It doesn't remove the need for clear metric rules. It makes those rules easier to apply again.

Save one monthly row with the four MRR movements. Add a note for unusual events, such as a price change or a large annual renewal. Over time, this becomes a useful movement log rather than a pile of disconnected snapshots.

Step 4: Segment Retention Before You Decide Whether It Is Good or Bad

A blended retention rate can hide the group causing the loss. Segment your SaaS retention benchmarks by customer type before you decide what needs fixing.

Start with customer size. Split accounts by ACV or monthly spend. Then add one more useful cut, such as acquisition channel, plan, region, or signup cohort.

A cohort is a group that started during the same period, usually the same month. Cohort retention asks how many customers from that group remain active at Month 1, Month 3, Month 6, or Month 12.

Suppose February had 200 new customers and January had only 50. February's total churn may look low because those new customers have not had much time to leave. A cohort view avoids that dilution.

Use rows for signup month and columns for months since signup. Then review the curve:

• A sharp early drop suggests weak onboarding or poor product fit.
• A curve that flattens after Month 3 suggests an early-fit problem.
• A steady decline points to ongoing value, support, or workflow issues.

Research on cohort analysis describes this as the difference between a point-in-time average and the behavior of a known customer group. For a deeper look at how to build these cohort views, see SaaS cohort analysis by signup month.

Next, compare activated and non-activated customers. Define activation as the first action that shows a customer reached value. The exact action depends on your product. It might be a connected data source, a completed report, or a first successful workflow.

Revenue cohorts add another layer. Track whether each cohort's MRR rises or falls after signup. A cohort can lose logos yet grow revenue if larger accounts expand. Another can retain many logos while losing revenue through downgrades.

Chartsy is useful when the question crosses systems. Its acquisition view can connect a UTM source to signups, customers, MRR, and later retention. That helps you spot a costly pattern: a channel that brings many trials but customers who churn early.

Step 5: Turn a Benchmark Gap Into a Retention Improvement Plan

A benchmark gap becomes useful only when it leads to one clear action. Start with the largest revenue leak, not the easiest metric to move.

If GRR is weak, focus on churn and contraction before chasing expansion. Read cancellation reasons by revenue, not just by customer count. Ten small cancellations may matter less than one large downgrade.

If GRR is stable but NRR is below target, inspect expansion. Look at which accounts have room to grow, then test a plan or usage path that matches their value. Don't push an upgrade to customers who haven't reached the first useful outcome.

If logo churn is high but revenue churn is low, your smaller accounts may be leaving while larger accounts stay. That can point to a pricing, onboarding, or customer-fit issue. It may also mean your blended logo number is less important than your revenue cohorts.

Failed payments need their own line. A customer whose card fails has a different problem from a customer who cancels after losing product value. Review payment failures before assigning all churn to customer success.

Build a 30-day plan with one owner and one expected metric change:

1. Find the segment with the largest lost MRR.
2. Read ten recent cancellation or downgrade records.
3. Choose one fix, such as better onboarding or a pause option.
4. Set a review date and compare the next cohort.

Track the result in Chartsy beside the original benchmark. If manual reporting is becoming a bottleneck, review the subscription analytics pricing while deciding whether to automate the workflow.

Keep acquisition in the same review. A high-volume channel may look good on signups but poor on retained MRR. Compare source, signup, paid conversion, expansion, and churn in one chain.

If you need to review product terms before connecting billing data, read the Terms of Use for Chartsy. Data access rules should be clear before the team builds a recurring reporting habit.

Don't set a target just because a report prints a high number. Set a target your segment can explain, then improve it through repeated cohort work.

FAQ: SaaS Retention Benchmarks

What is a good SaaS retention rate?

A good rate depends on the metric, stage, and customer segment. Current benchmark sources report median NRR near 101%, median GRR near 84%, and monthly churn near 3.5%. Use those as context, then compare the same metric with your own past cohorts.

What is a good NRR for a small SaaS?

For a small SaaS, NRR near or above 100% usually means expansion offsets churn and downgrades. But SMB products often have lower expansion room than enterprise products. Compare NRR with GRR, since a strong upsell motion can hide customer loss.

How do I calculate SaaS retention?

Calculate SaaS retention by fixing a starting customer group, then measuring its revenue or logo count at the end of the period. For NRR, add expansion and subtract contraction plus churn. Divide by starting recurring revenue and multiply by 100.

What is the difference between NRR and GRR?

NRR includes expansion from existing customers, while GRR excludes it. NRR can rise above 100% when upgrades outweigh losses. GRR cannot exceed 100%, so it gives a cleaner view of churn and downgrades inside the starting base.

Should I track monthly or annual churn?

Track both when your data volume allows it. Monthly churn shows early movement, while annual churn shows the compounded effect across a full customer cycle. Never compare the two without labeling the period, customer group, and churn definition.

Conclusion

Use retention benchmarks as a map, not a verdict. Pick one metric, compare it with a matching segment, then trace the result back to customer and revenue movements. Start with one clean cohort this month. Chartsy can help you bring Stripe or Paddle data into one view, but the next decision still belongs to you: choose the largest retention gap and test one fix.