Retention Metrics
Understand the difference between Gross Revenue Retention (GRR) and Net Revenue Retention (NRR). Why GRR is always ≤100%, how NRR captures expansion, and benchmarks for each.
GRR (Gross Revenue Retention) measures the percentage of recurring revenue retained from existing customers, excluding any expansion revenue. It only accounts for revenue lost to churn (customers leaving entirely) and contraction (customers downgrading). GRR is always less than or equal to 100% because it cannot benefit from upsells or cross-sells. It is the purest measure of your ability to keep the revenue you already have.
NRR (Net Revenue Retention), also called Net Dollar Retention (NDR), measures the percentage of recurring revenue retained from existing customers including expansion revenue from upsells, cross-sells, and seat additions. NRR can exceed 100% when expansion outpaces losses, a phenomenon called net negative churn. NRR shows the total revenue trajectory of your existing customer base.
The critical difference: GRR answers "how much revenue are we losing?" while NRR answers "is our existing customer base growing or shrinking?" A company with low GRR but high NRR is losing many customers while aggressively expanding the ones that stay. A company with high GRR but moderate NRR retains well but under-monetizes its base.
<h2>The Detailed Comparison</h2>
<table>
<thead>
<tr>
<th>Dimension</th>
<th>GRR (Gross Revenue Retention)</th>
<th>NRR (Net Revenue Retention)</th>
</tr>
</thead>
<tbody>
<tr><td><strong>Definition</strong></td><td>Revenue retained excluding expansion</td><td>Revenue retained including expansion</td></tr>
<tr><td><strong>Includes expansion revenue</strong></td><td>No</td><td>Yes</td></tr>
<tr><td><strong>Maximum value</strong></td><td>100%</td><td>No cap (commonly 80%–150%+)</td></tr>
<tr><td><strong>What it reveals</strong></td><td>Retention quality and product stickiness</td><td>Net revenue growth from existing customers</td></tr>
<tr><td><strong>Best-in-class benchmark</strong></td><td>Above 90%</td><td>Above 120%</td></tr>
<tr><td><strong>Concerning threshold</strong></td><td>Below 80%</td><td>Below 100%</td></tr>
<tr><td><strong>Primary audience</strong></td><td>Product, CS, and retention teams</td><td>Executives, investors, and board members</td></tr>
<tr><td><strong>Strategic signal</strong></td><td>Product-market fit and customer satisfaction</td><td>Revenue compounding power and expansion efficiency</td></tr>
<tr><td><strong>Segment sensitivity</strong></td><td>Enterprise GRR is typically higher than SMB</td><td>Enterprise NRR is typically higher due to larger expansion potential</td></tr>
</tbody>
</table>
<h2>Formulas</h2>
<div class="not-prose mt-4 mb-6 space-y-3">
<div class="rounded-lg bg-gradient-to-r from-primary/5 to-primary/[0.02] border border-primary/10 px-5 py-3.5 font-mono text-sm">
GRR = (Starting MRR − Contraction MRR − Churned MRR) ÷ Starting MRR × 100
</div>
<div class="rounded-lg bg-gradient-to-r from-primary/5 to-primary/[0.02] border border-primary/10 px-5 py-3.5 font-mono text-sm">
NRR = (Starting MRR + Expansion MRR − Contraction MRR − Churned MRR) ÷ Starting MRR × 100
</div>
</div>
<p>The only difference in the formulas is that NRR adds Expansion MRR in the numerator. This single term is what allows NRR to exceed 100%. GRR can only subtract from the starting base, so it is structurally capped at 100%.</p>
<h2>Real-World Example</h2>
<p>Consider a SaaS company with the following monthly MRR movements:</p>
<ul>
<li><strong>Starting MRR:</strong> $200,000</li>
<li><strong>Expansion MRR (upsells, seat additions):</strong> $18,000</li>
<li><strong>Contraction MRR (downgrades):</strong> $6,000</li>
<li><strong>Churned MRR (cancellations):</strong> $8,000</li>
</ul>
<p><strong>GRR:</strong> ($200,000 − $6,000 − $8,000) ÷ $200,000 × 100 = <strong>93%</strong>. The company retains 93% of its starting revenue before accounting for expansion. This is a healthy result indicating strong product-market fit.</p>
<p><strong>NRR:</strong> ($200,000 + $18,000 − $6,000 − $8,000) ÷ $200,000 × 100 = <strong>102%</strong>. After including expansion, the existing customer base actually grew by 2%. This means the company can grow revenue even without acquiring a single new customer.</p>
<p>Now consider what happens if expansion drops to $5,000 while losses stay the same:</p>
<p><strong>GRR:</strong> Still 93% (expansion does not affect GRR). <strong>NRR:</strong> ($200,000 + $5,000 − $6,000 − $8,000) ÷ $200,000 × 100 = <strong>95.5%</strong>. NRR drops below 100%, meaning the existing base is now shrinking. The company must acquire new revenue just to stand still.</p>
<h2>Industry Benchmarks</h2>
<p>These benchmarks are directional. For current data by company stage and segment, refer to annual SaaS surveys from <a href="https://openviewpartners.com" target="_blank" rel="noopener noreferrer">OpenView</a> and <a href="https://chartmogul.com" target="_blank" rel="noopener noreferrer">ChartMogul</a>.</p>
<ul>
<li><strong>Median GRR (B2B SaaS):</strong> 85–90%. Enterprise-focused companies tend to cluster at 90–95%, while SMB-focused companies often fall in the 75–85% range due to higher logo churn in the small business segment.</li>
<li><strong>Median NRR (B2B SaaS):</strong> 100–110%. Top-quartile public SaaS companies consistently report 115–130% NRR, driven by strong expansion motions and usage-based pricing.</li>
<li><strong>Best-in-class examples:</strong> Companies like Snowflake, Datadog, and Twilio have historically reported NRR above 130%, demonstrating powerful land-and-expand models where customers dramatically increase usage over time.</li>
<li><strong>Red flag:</strong> GRR below 75% combined with NRR below 100% indicates both poor retention and insufficient expansion. This pattern is difficult to sustain because you are losing customers faster than you can grow the survivors.</li>
</ul>
<h2>When to Use Each</h2>
<p><strong>Use GRR when:</strong></p>
<ul>
<li>You want to evaluate the core stickiness of your product without expansion masking retention problems</li>
<li>You are diagnosing whether churn is a product problem, pricing problem, or customer success problem</li>
<li>You are comparing retention across customer segments (SMB vs mid-market vs enterprise)</li>
<li>You are assessing the durability of your revenue base in a downside scenario where expansion slows</li>
<li>You are benchmarking against peers in your segment, since GRR is less affected by pricing model differences</li>
<li>You are evaluating product-market fit for a specific customer cohort</li>
</ul>
<p><strong>Use NRR when:</strong></p>
<ul>
<li>You are reporting to the board or investors about the overall health of your customer base</li>
<li>You want to understand the revenue growth rate your existing customers contribute without new acquisition</li>
<li>You are evaluating the effectiveness of your expansion, upsell, and cross-sell motions</li>
<li>You are building a revenue forecast that accounts for both retention and expansion</li>
<li>You are assessing whether your company can achieve net negative churn</li>
<li>You are comparing your growth efficiency against public SaaS benchmarks</li>
</ul>
<h2>Common Mistakes</h2>
<p><strong>Mistake 1: Tracking NRR but ignoring GRR.</strong> NRR can look healthy while GRR deteriorates, because expansion masks churn. If GRR drops from 90% to 75% but NRR stays at 110%, you have a growing retention problem that expansion is temporarily hiding. When expansion eventually slows (and it always does during downturns), the underlying churn becomes visible and painful.</p>
<p><strong>Mistake 2: Mixing up logo retention with revenue retention.</strong> Logo retention counts the percentage of customers retained regardless of spend. Revenue retention weighs each customer by their revenue. Losing one $50K enterprise customer has a dramatically different revenue retention impact than losing ten $500 SMB customers, even though logo retention treats them the same. Always clarify whether you are discussing logo or revenue retention.</p>
<p><strong>Mistake 3: Including new customers in the retention calculation.</strong> Both GRR and NRR measure revenue from existing customers only. Revenue from new customers acquired during the period must be excluded. If you add new customer revenue to the numerator, you inflate retention metrics and lose the signal about how well you serve your installed base.</p>
<p><strong>Mistake 4: Using inconsistent time periods.</strong> GRR and NRR should be calculated on the same cohort over the same time period. Monthly calculations reveal trends faster but can be noisy. Annual calculations are smoother but slower to signal problems. Pick one cadence for operational tracking (monthly) and another for external reporting (quarterly or annually), but be consistent.</p>
<p><strong>Mistake 5: Treating GRR and NRR as independent metrics.</strong> GRR and NRR are most powerful when analyzed together. The gap between them reveals your expansion efficiency: NRR − GRR = the expansion contribution. If GRR is 88% and NRR is 115%, expansion is contributing 27 percentage points. If that expansion dries up, NRR drops to 88% immediately.</p>
<h2>How They Connect to Other Metrics</h2>
<ul>
<li><strong>GRR drives LTV floor:</strong> GRR sets a floor for customer lifetime value. If GRR is 80%, your average customer loses 20% of their revenue annually, which directly reduces LTV compared to a company with 95% GRR.</li>
<li><strong>NRR drives ARR growth:</strong> NRR above 100% means your existing base contributes positive growth to ARR even without new bookings. A company with $10M ARR and 120% NRR adds $2M from expansion alone, reducing pressure on new logo acquisition.</li>
<li><strong>Both connect to CAC payback:</strong> Higher retention (GRR) means customers stay longer, recovering CAC over more months. Higher expansion (NRR) means customers pay more over time, accelerating effective payback.</li>
<li><strong>GRR reveals churn impact:</strong> GRR is the complement of your gross churn rate. If GRR is 93%, your gross monthly revenue churn is 7%. Use the Churn Impact calculator to model how that compounds over time.</li>
<li><strong>NRR powers the Magic Number:</strong> Strong NRR reduces the burden on new customer acquisition, improving Magic Number by ensuring that ARR growth comes from both new and existing revenue sources.</li>
</ul>
<h2>Sources and Benchmarks</h2>
<p>Benchmarks cited are directional and drawn from publicly available SaaS survey data. For current GRR and NRR benchmarks by stage and segment, refer to annual reports from <a href="https://openviewpartners.com" target="_blank" rel="noopener noreferrer">OpenView</a> and <a href="https://chartmogul.com" target="_blank" rel="noopener noreferrer">ChartMogul</a>. Always validate against your own cohort data.</p>
<h2>The Bottom Line</h2>
<p>GRR and NRR are two sides of the same retention coin. GRR tells you how well you keep revenue, isolating the losses. NRR tells you the net result after expansion offsets those losses. Best-in-class SaaS companies achieve GRR above 90% and NRR above 120%, meaning they retain almost all existing revenue and grow it significantly through expansion.</p>
<p>Track both. If GRR is declining, focus on churn reduction and product stickiness before investing more in expansion. If GRR is strong but NRR is flat, invest in upsell motions, usage-based pricing, and cross-sell opportunities. The companies that win on both metrics build the most durable, capital-efficient growth engines.</p>Yes, and it happens more often than you might think. If expansion revenue from upsells and cross-sells is large enough to offset churn and contraction, NRR can exceed 100% even when GRR is relatively low. For example, a company with 75% GRR but aggressive land-and-expand motions might achieve 115% NRR. However, this pattern is fragile: it depends on continuous expansion from a shrinking base of retained revenue. Investors will flag low GRR even when NRR looks healthy because it signals underlying retention problems.
NRR typically carries more weight in SaaS valuations because it captures the full picture of customer revenue dynamics, including expansion. Public SaaS companies with NRR above 120% trade at significantly higher multiples than those below 100%. However, GRR matters during due diligence: a company with 130% NRR but 70% GRR will face harder questions than one with 115% NRR and 90% GRR. The best-valued companies excel at both.
Calculate both monthly for internal tracking and report them quarterly to the board and investors. Monthly tracking reveals trends and seasonal patterns early. Quarterly reporting smooths out noise and aligns with standard SaaS reporting cadences. For cohort-level analysis, calculate GRR and NRR by customer segment (SMB, mid-market, enterprise) to identify where retention is strongest and weakest.
Yes, they describe the same phenomenon from different angles. Net negative churn means expansion revenue from existing customers exceeds the revenue lost to churn and contraction. When this happens, NRR exceeds 100%. A company with 3% gross revenue churn but 5% expansion has net negative churn of -2%, which translates to 102% NRR. The terms are interchangeable but NRR is the more standardized reporting metric.