NRR above 100% is a fine goal for a high-growth SaaS business, but for most CS teams managing 200 to 1,000 mid-market accounts, it's not a useful operational benchmark. It conflates expansion revenue — which is partly a sales motion — with retention performance, which is what your CS program actually controls. A CS team with flat expansion but zero logo churn is doing something fundamentally different from one with modest logo churn but aggressive upsell. Both might land above 100% NRR; both might land below it. The number alone doesn't tell you much about your retention program's health.
This piece is about the benchmarks that actually tell you whether your CS program is above or below par — broken down by the retention metrics that are operationally meaningful at mid-market scale.
The Metrics That Matter: A Hierarchy
Before benchmarks, a quick taxonomy. These terms get conflated constantly in CS reporting, and the conflation produces decisions based on the wrong numbers.
- Logo churn rate: the percentage of accounts (logos) that don't renew in a period, regardless of their contract value. A 5% logo churn rate means 5 of every 100 accounts didn't renew.
- Revenue churn rate (gross): the ARR lost from non-renewals and contractions as a percentage of ARR at the start of the period. Does not include expansion.
- Net Revenue Retention (NRR): the revenue retained from the existing customer base after accounting for churn, contraction, and expansion. Above 100% means expansion is outpacing loss; below 100% means the installed base is shrinking.
- Gross Revenue Retention (GRR): NRR with expansion removed — it measures only churn and contraction. GRR is a cleaner measure of your retention program's performance because it isolates the loss side from the growth side.
For mid-market CS benchmarking, GRR is the number to track. It's the most direct measure of what your CS program controls: are you keeping the accounts you already have?
What Good Actually Looks Like: Benchmarks by Segment
Industry benchmarks for B2B SaaS retention vary by average contract value (ACV), product category, and company stage. The following ranges are drawn from patterns visible across the CS industry and should be treated as directional rather than exact.
Logo Churn Rate — Annual
For mid-market B2B SaaS (ACV $5K–$50K), annual logo churn benchmarks generally fall in these ranges:
- Strong performance: 3–5% annual logo churn
- Median performance: 6–10% annual logo churn
- Underperforming: Above 12% annual logo churn
At 5–7% annual logo churn, you're losing roughly 1 account per 15–20 managed. At 12%, you're turning over more than 1 in 8 accounts every year — which means you're on a growth treadmill, constantly replacing churn before you can show net expansion.
Gross Revenue Retention — Annual
- Strong performance: 90%+ GRR
- Median performance: 83–89% GRR
- Underperforming: Below 80% GRR
90%+ GRR is achievable for mid-market B2B SaaS with strong CS programs and relatively sticky products. Sub-80% GRR is almost always a signal of either poor product-market fit, significant service quality issues, or both — it's rarely a pure CS execution problem.
Net Revenue Retention — Annual
- Strong performance: 105–120% NRR (expansion outpacing churn)
- Median performance: 95–104% NRR
- Underperforming: Below 90% NRR
Worth noting: many very healthy mid-market SaaS businesses run at 95–103% NRR. That doesn't represent an expansion-story-worthy number, but for a CS team managing 200–500 accounts with a modest upsell motion, it reflects solid core retention work.
The Benchmark Traps to Avoid
We're not saying high NRR is a bad signal — for businesses with genuine net expansion from their installed base, it reflects real value delivery. We're saying that benchmarking NRR alone, without decomposing GRR, can lead CS teams to optimize for the wrong variable.
The most common benchmark trap is "our NRR looks fine" masking a GRR problem. If you're at 102% NRR but only 81% GRR, your expansion revenue is compensating for a logo churn problem that will compound over time. When you run out of low-tier accounts to upsell into the mid-tier, your NRR will fall and you'll be staring at a structural retention problem that's been hiding behind expansion.
A second trap: comparing your retention numbers to public SaaS benchmarks from high-ACV enterprise businesses. Companies with $100K+ average contracts have very different churn dynamics (longer sales cycles, more embedded workflows, more executive-level relationship investment) than mid-market tools at $10K–$30K ACV. Comparing your 8% logo churn to a $200K ACV enterprise software company's 3% churn is comparing different categories.
A Practical Scenario: Reading Your Cohort Data
Imagine you're running CS for a field service management SaaS — 210 accounts, average ACV $18K, two-person CS team. Your retention numbers for 2024: GRR 86%, NRR 94%, logo churn 9%.
Against the benchmarks above: GRR at 86% is solidly median. Logo churn at 9% is at the higher edge of median. NRR at 94% is slightly below the median range. That's a picture of a CS program that is keeping most of its revenue but losing accounts slightly faster than the benchmark for a healthy mid-market program. The NRR gap isn't an expansion problem — it's a retention problem.
What does the cohort breakdown tell you? Pull the 19 accounts that churned. Were they disproportionately in a specific tier, product line, or geographic cluster? Were they accounts where the CSM had low logged activity in the 90 days before renewal? Did support ticket volume spike before the churn event? Each of these diagnostic questions maps back to something operational your team can change.
The Cohort Analysis Imperative
Aggregate retention metrics are a lagging indicator. By the time GRR drops from 88% to 83%, the churn pattern that caused it is already 6–12 months old. Cohort analysis — tracking retention rates by account acquisition quarter, by product tier, by CSM, by ICP segment — surfaces the pattern before it aggregates into an alarming annual number.
For CS teams using Gainsight, Salesforce, or any CRM with decent reporting, cohort analysis should be a quarterly practice. The questions to answer:
- Are accounts acquired in recent cohorts retaining better or worse than older cohorts at the same age?
- Do accounts onboarded by one CSM have meaningfully different 12-month retention rates than accounts onboarded by another?
- Does time-to-value (days to first meaningful usage milestone) correlate with 12-month retention in your data?
If accounts with time-to-value under 14 days retain at 93% versus accounts with time-to-value over 30 days retaining at 79%, you've found a lever. Onboarding quality is directly controlling a meaningful share of your retention outcome.
What the Benchmarks Don't Tell You
Being at the median isn't a good enough answer for a CS team trying to improve. The benchmark tells you where you stand relative to the market. It doesn't tell you what you're capable of with better signal coverage, better playbook execution, or better renewal forecasting accuracy.
The teams consistently at the 90%+ GRR mark in mid-market B2B SaaS share a few common traits: they have early warning signals that fire 60–90 days before renewal rather than at renewal time, they have CSM playbooks that execute automatically rather than depending on manual review, and they have renewal forecasts accurate enough to allocate CSM attention to the accounts that actually need it.
Those are operational choices, not luck. The benchmark just tells you what the gap looks like from where you're standing.