Decompose net revenue retention into the drivers that move it
Net revenue retention: formula, benchmarks and levers
Net revenue retention is the single best indicator of whether a SaaS business can compound growth from its existing customer base. An NRR above 100% means your installed base is growing on its own, without a single new logo. But NRR is a composite outcome. It is the net result of gross retention, expansion revenue, and contraction. To move NRR deliberately, you need to decompose it into the operational levers underneath. This guide walks through the NRR formula, shows how to build a metric tree that decomposes it, shares benchmarks by segment and stage, and identifies the levers that reliably improve each branch.
9 min read
What is net revenue retention?
Net revenue retention (NRR), sometimes called net dollar retention (NDR), measures the percentage of recurring revenue retained from existing customers over a given period, after accounting for expansion, contraction, and churn. It answers a deceptively simple question: if you stopped acquiring new customers today, would your revenue grow, stay flat, or shrink?
The formula is straightforward:
NRR = (Beginning MRR + Expansion MRR - Contraction MRR - Churned MRR) / Beginning MRR
The result is expressed as a percentage. An NRR of 110% means that for every pound of recurring revenue you started with, you now have a pound and ten pence, purely from existing customers. An NRR of 90% means you are losing ten pence on every pound, and your acquisition engine must fill that gap before you can grow.
NRR is typically calculated on a monthly basis using MRR, then annualised or reported as a trailing twelve-month figure for comparability. Some companies calculate it directly from ARR. The mechanics are the same; only the time window changes.
What makes NRR powerful is that it captures three forces in a single number. Gross retention tells you how well you hold onto revenue. Expansion tells you how much additional revenue you generate from customers who stay. Contraction tells you how much revenue you lose from downgrades even when customers do not leave entirely. NRR nets all three together. It is the truest measure of the value your product delivers to customers who have already chosen to pay for it.
NRR captures three forces in one number: how well you retain revenue (gross retention), how much you grow within existing accounts (expansion), and how much you lose to downgrades (contraction). An NRR above 100% means your customer base is a self-sustaining growth engine.
Why NRR matters for SaaS valuation and growth
NRR has become the metric that investors scrutinise most closely when evaluating SaaS businesses. The reason is mathematical: NRR determines the long-term revenue trajectory of every cohort you acquire. A company with 120% NRR will see each cohort nearly triple its starting revenue within six years. A company with 90% NRR will see each cohort lose more than half its starting revenue over the same period. The compounding effect is dramatic, and it shows up directly in enterprise value.
Public SaaS companies with NRR above 120% consistently trade at higher revenue multiples than their peers. Companies like Snowflake, Datadog, and Twilio built enormous market capitalisations in large part because their usage-based models drove NRR well above 130%. Investors pay a premium for NRR because it de-risks the growth story. A company with high NRR does not need its sales team to perform heroically every quarter just to maintain the current revenue run rate. The existing base provides a rising floor.
Beyond valuation, NRR shapes the economics of the entire business. When NRR exceeds 100%, each pound spent on customer acquisition generates a stream of revenue that grows over time rather than decaying. This means the lifetime value of a customer increases, the CAC payback period shortens, and the LTV:CAC ratio improves. High NRR businesses can afford to invest more aggressively in acquisition because the returns compound. Low NRR businesses face the opposite dynamic: every customer they acquire is a depreciating asset, and they must run faster just to stand still.
| NRR level | What it signals | Typical business characteristics |
|---|---|---|
| Below 80% | Severe retention problem | High churn, weak product-market fit, or serving a segment with inherently high turnover. Revenue base is shrinking rapidly and acquisition cannot keep pace indefinitely. |
| 80% to 95% | Revenue leakage | Common in SMB-focused SaaS where customer turnover is naturally higher. Sustainable if acquisition cost is low, but limits valuation multiples and long-term compounding. |
| 95% to 105% | Stable but not compounding | The customer base is roughly holding steady. Expansion and contraction are roughly balanced. Growth depends entirely on new logo acquisition. |
| 105% to 120% | Healthy expansion | Strong indication that customers are finding increasing value. The installed base is a growth engine. Common in mid-market and enterprise SaaS with upsell and seat expansion motions. |
| Above 120% | Exceptional compounding | Best-in-class. Typically seen in usage-based or platform businesses where customer spend scales with their own growth. Drives premium valuations and efficient growth. |
The lesson for SaaS leaders is that NRR is not just a customer success metric. It is a company-level strategic metric that determines the efficiency of your growth model, the return on your acquisition spend, and ultimately the enterprise value of the business. Treating NRR as a number that only the customer success team owns is a common and costly mistake.
Decomposing NRR with a metric tree
NRR is a composite metric. To improve it, you need to decompose it into the components that drive it. A metric tree makes this decomposition visual and actionable by showing the causal relationships between NRR and its operational drivers.
At the first level, NRR decomposes into three branches: gross revenue retention (GRR), expansion rate, and contraction rate. GRR measures the percentage of revenue retained before accounting for any expansion. It is the floor beneath NRR. Expansion rate measures the percentage of beginning revenue added through upsells, cross-sells, and seat growth. Contraction rate measures the percentage lost to downgrades and seat reductions without full churn.
The relationship is: NRR = GRR + Expansion Rate - Contraction Rate.
Each of these branches decomposes further into operational drivers that specific teams can influence.
This tree reveals several things that a single NRR number hides.
First, it separates gross retention from expansion. Two companies can both report 110% NRR, but one might have 95% GRR with 15% expansion, while the other has 85% GRR with 25% expansion. The first company has a healthier foundation. The second is masking a significant churn problem with aggressive upselling, and that strategy has a ceiling.
Second, it distinguishes voluntary churn from involuntary churn. Voluntary churn (customers actively deciding to leave) requires product improvements, better onboarding, or competitive repositioning. Involuntary churn (failed payments) requires better billing infrastructure, card retry logic, and dunning workflows. The causes are entirely different, so the solutions must be too.
Third, it breaks expansion into its distinct motions. Upsells (moving customers to higher tiers), cross-sells (selling additional products), and seat expansion (adding users) are driven by different teams and different product capabilities. Lumping them together under "expansion" obscures which motions are working and which are stalling.
Fourth, it makes contraction visible as a separate force. Many companies track net churn (churn minus expansion) but do not separately track contraction. Contraction is a distinct signal: customers are staying but finding less value, or at least less value at their current price point. Understanding why customers downgrade provides different insights from understanding why they leave entirely.
NRR benchmarks by segment and stage
What counts as "good" NRR depends heavily on your customer segment, contract structure, and company stage. Comparing your NRR to an irrelevant benchmark leads to either complacency or unnecessary alarm. The benchmarks below reflect data aggregated from public company filings, industry surveys, and investor frameworks.
SMB-focused SaaS
Median NRR of 90% to 100%. SMB customers churn at higher rates due to business closures, budget constraints, and lower switching costs. NRR above 100% is achievable but requires strong seat expansion or usage-based pricing. Gross retention typically sits between 80% and 90%.
Mid-market SaaS
Median NRR of 100% to 110%. Longer contracts and deeper integrations reduce churn. Expansion comes from seat growth as customer organisations scale and from upsells to higher tiers. Gross retention typically sits between 90% and 95%.
Enterprise SaaS
Median NRR of 110% to 130%. Multi-year contracts, high switching costs, and large account expansion drive strong retention and growth. Best-in-class enterprise SaaS companies sustain NRR above 125%. Gross retention often exceeds 95%.
Usage-based SaaS
Median NRR of 115% to 140%+. Revenue scales with customer usage, creating natural expansion without requiring explicit upsell conversations. Companies like Snowflake and Datadog have reported NRR above 130%. However, usage-based models can also amplify contraction during downturns.
Early-stage (£1M-£10M ARR)
Median NRR around 95% to 105%. Early-stage companies are still refining product-market fit and expansion motions. NRR below 90% at this stage is a warning sign that the core value proposition is not landing. The priority should be gross retention before expansion.
Scale-up (£30M+ ARR)
Median NRR of 105% to 115%. At scale, NRR tends to moderate as the customer base diversifies and the easiest expansion opportunities have been captured. Maintaining NRR above 110% at scale is a sign of strong product-market fit and effective account management.
GRR matters as much as NRR
Gross revenue retention (GRR) is the floor beneath NRR. GRR removes expansion from the picture and shows how well you hold onto the revenue you already have. Best-in-class SaaS companies maintain GRR above 90%. If your GRR is below 85%, no amount of expansion will create a sustainable growth engine. Fix retention before optimising expansion.
Levers to improve NRR
Improving NRR is not a single initiative. It requires coordinated action across the three branches of the metric tree: reducing churn to raise GRR, driving expansion, and minimising contraction. The levers below are organised by branch, so teams can identify which area offers the most headroom and focus their efforts accordingly.
- 1
Strengthen onboarding to reduce early churn
Most churn is decided in the first 90 days. Customers who reach a meaningful activation milestone within the first two weeks retain at dramatically higher rates. Map your product to the "aha moment" and design onboarding to reach it as quickly as possible. Track time-to-value as a leading indicator of gross retention. Every day of confusion or delay is a day closer to cancellation.
- 2
Build health scoring to predict churn before it happens
By the time a customer submits a cancellation request, the decision was made weeks ago. Customer health scores that combine product usage data (login frequency, feature adoption depth, support ticket sentiment) with commercial signals (contract end date, stakeholder changes) give customer success teams time to intervene. The goal is to move from reactive saves to proactive value delivery.
- 3
Fix involuntary churn with payment infrastructure
Involuntary churn from failed payments typically accounts for 20% to 40% of total churn in SaaS businesses. Smart retry logic (retrying charges at optimal times), pre-dunning emails before card expiry, and frictionless card update flows can recover a significant portion of this revenue. It is the highest-ROI retention investment most companies overlook.
- 4
Design pricing that rewards growth
Usage-based and seat-based pricing models create natural expansion as customers grow. When expansion is built into the pricing structure rather than requiring a sales conversation, it happens automatically. Align your pricing metric with the unit of value the customer receives, so that as they get more value, you capture more revenue. This is the single most powerful lever for NRR above 120%.
- 5
Invest in product capabilities that drive upsells
Expansion happens when customers discover new problems your product can solve. Build features that serve adjacent use cases and package them in higher tiers. Track feature adoption by tier to identify which capabilities are most likely to trigger upgrades. The product roadmap should explicitly include features designed to drive expansion, not just features designed to prevent churn.
- 6
Reduce contraction by aligning value to pricing tiers
Customers downgrade when they feel they are paying for value they are not receiving. Audit your tier structure to ensure each plan level delivers a clear, differentiated set of capabilities that justifies the price difference. When customers downgrade, conduct exit interviews to understand the gap. Often, contraction signals a packaging problem rather than a product problem.
“The most capital-efficient growth comes from expanding existing customers, not acquiring new ones. A pound invested in expansion typically returns three to five times more than a pound invested in acquisition, because the customer relationship and trust already exist.”
Connecting NRR to other SaaS metrics
NRR does not exist in isolation. It connects to and influences nearly every other metric in a SaaS business. Understanding these connections is essential for building a metric tree that reflects how the business actually works, rather than a collection of siloed numbers.
NRR and LTV are directly linked. Lifetime value is a function of how long a customer stays and how much they pay over that period. NRR above 100% means customer revenue grows over time, which increases LTV without increasing acquisition cost. This is why NRR has such a powerful effect on LTV:CAC ratios. A five-percentage-point improvement in NRR can improve LTV by 30% or more over a multi-year horizon.
NRR and CAC payback interact through the speed of revenue recovery. When NRR is high, each cohort generates more revenue in its second year than its first, which accelerates the payback on the initial acquisition cost. A company with 120% NRR might achieve full CAC payback in 12 months, while a company with 90% NRR and the same starting ACV might take 24 months.
NRR and ARR growth rate are connected through the composition of growth. Total ARR growth comes from two sources: new logo ARR and net expansion from existing customers. A company growing ARR at 50% with 120% NRR has a fundamentally different growth profile from a company growing at 50% with 90% NRR. The first gets a significant portion of its growth "for free" from the installed base. The second must generate all its growth from new customer acquisition, which is more expensive and harder to sustain.
| Metric | How NRR connects | Why the connection matters |
|---|---|---|
| Lifetime value (LTV) | NRR above 100% increases revenue per customer over time, directly raising LTV | Higher LTV improves unit economics and justifies greater acquisition investment |
| CAC payback period | Stronger NRR accelerates revenue recovery, shortening the time to recoup acquisition cost | Shorter payback frees capital for reinvestment and reduces the risk of customer loss before breakeven |
| Gross margin | Expansion revenue often comes at higher margins than new revenue because onboarding and implementation costs are already incurred | High NRR improves blended gross margins, increasing the cash-flow efficiency of every revenue pound |
| Rule of 40 | High NRR contributes to growth rate while reducing the spend needed to sustain it, improving the growth-plus-margin equation | Companies with strong NRR can achieve the Rule of 40 with less aggressive spending, which investors favour |
| Burn multiple | NRR-driven growth requires less cash than acquisition-driven growth, lowering the burn multiple | A lower burn multiple signals efficient growth, which is increasingly important in capital-constrained environments |
The practical implication is that NRR should sit prominently in your company-level metric tree, not buried in a customer success dashboard. It connects upward to ARR and valuation. It connects sideways to CAC, LTV, and unit economics. It connects downward to product engagement, onboarding quality, and pricing strategy. When you build a metric tree with NRR as a key node, you make all of these connections visible, and you give every team a clear line of sight from their daily work to the company-level outcomes that matter.
KPI Tree is built for exactly this kind of cross-functional decomposition. It lets you model NRR as a node in your metric tree, connect it to live data from your billing and product analytics systems, assign ownership to the teams that influence each branch, and track the actions they take to move the numbers. The result is a living system where NRR is not a quarterly report but a real-time signal that drives daily decisions.
Continue reading
Metric trees for SaaS companies
Decomposing recurring revenue into the levers that drive it
Customer acquisition cost: a metric tree decomposition
Decompose CAC into its component parts so you can see exactly where your acquisition spend goes and how to improve it
Metric decomposition
Break any business metric into the components that drive it
Decompose NRR into the levers that drive it
Build a metric tree that connects net revenue retention to gross retention, expansion, and contraction. Assign ownership, connect to live data, and track the actions that move your numbers.
