Net revenue retention, explained simply

2026-08-04 · By Lubin Danilo, founder of Lubin Investment

Net revenue retention measures how much existing customers spend from one year to the next, excluding new customers entirely. Above 100%, existing customers are spending more than before: it is a signal that a product is becoming more essential over time, not less. That is why this single number carries so much weight in how software companies get valued.

The problem this number solves

A company announcing 20% revenue growth over a year sounds healthy. But that sentence alone says nothing about the real question that matters for judging how solid that growth is: where does it come from? If the 20% comes entirely from new customers won through heavy advertising and sales spending, while existing customers spend less and less or leave in droves, that company is running on a treadmill that keeps accelerating: it must spend ever more just to offset an eroding base underneath. If instead a good chunk of that 20% comes from customers who were already there the year before buying more, without even counting new customers, that is a radically different story: the product is becoming more essential over time, not less.

This is exactly what total revenue growth, taken alone, cannot distinguish. Two companies can show the same 20% growth with opposite stories underneath: one building on sand by constantly recruiting to offset departures, the other building on rock because its existing customers deepen their usage year after year. An investor, or more simply anyone trying to judge whether a software company deserves its valuation, needs a number that separates these two stories. That is the role of net revenue retention.

This problem is especially acute for subscription software companies, because their entire business model rests on the length of the customer relationship. A company selling a physical object once does not face this problem the same way: the sale is done, the customer walks away with the product. A subscription software vendor, on the other hand, must convince the same customer to keep paying every month or every year, indefinitely, and ideally to pay a little more each year because they use the product more, add users, or upgrade. The entire long-term value of this type of business rests on that capacity to retain and grow existing customers, far more than on the pace of acquiring new ones.

How it is calculated, and what it actually captures

The calculation, in principle, is simple: take the revenue generated by a specific group of customers at a given point in time, say a year ago, and measure how much that same group, the exact same companies or accounts, spends today. Any new customer acquired in between is deliberately excluded: this number speaks only to the behavior of customers who were already there. If that group spent $100 million a year ago and spends $120 million today, net revenue retention is 120%. If it spends $90 million today, the rate is 90%.

This number actually aggregates three opposing movements within the same customer group. On one side, expansion: existing customers adding users, buying an extra module, or upgrading to a pricier tier because their own business is growing and they use the product more. On the other side, two forms of loss: contraction, when a customer reduces usage or budget without fully leaving (they lay off staff, cut the number of seats), and churn, literally the customer's complete departure, no longer paying at all. A net revenue retention rate above 100% means expansion clearly outweighs contraction and churn combined: gains exceed losses within the same customer base, without even counting the contribution of new customers.

A rate below 100% does not necessarily mean an imminent disaster, but it is a signal worth watching, especially if it persists over several consecutive quarters. Available market data on private software companies shows the median rate sliding from around 105% in 2021 to around 101% in 2024 (SaaS market data), a clear slowdown in expansion dynamics across the entire sector, in a context where corporate IT budgets have come under more scrutiny than at the peak of the post-pandemic cycle. The impact on public market valuations is striking: publicly traded software companies above 120% net retention have traded at roughly 9.3 times revenue, versus roughly 3.1 times for those below 100%, a threefold valuation gap that shows just how much this single number weighs in judging the quality of a subscription model.

Two real companies that show the gap

ServiceNow, the software vendor that automates internal enterprise processes (IT incident management, employee requests, financial approvals), consistently reports net revenue retention above 120%. The mechanism behind that number is not a sales trick, it is structural lock-in: once its platform is embedded in a large enterprise's daily processes, with years of specific configurations, staff training, and automations built on top, switching vendors becomes extremely costly and risky for the customer itself. This lock-in does not just keep customers from leaving: it gives them a rational reason to add more modules and users as their own company grows, rather than fragmenting their tools across several competing vendors.

Datadog, the software vendor that monitors IT infrastructure performance (servers, applications, databases) to catch failures before they hit end users, illustrates a different but equally powerful expansion mechanism. Its trailing-twelve-month net revenue retention sat in the low 120s in the first quarter of 2026, up from the high 110s a year earlier (Datadog quarterly report). Here, expansion comes directly from usage-based billing: the bigger a customer's infrastructure grows, the more data there is to monitor, and the bigger the Datadog bill grows mechanically, with no salesperson needing to renegotiate a contract. The product literally grows at the same pace as the customer's own business, a dynamic that also explains why Datadog crossed the billion-dollar mark in quarterly revenue for the first time.

These two mechanisms, switching-cost lock-in at ServiceNow and usage-proportional billing at Datadog, are the two big families of causes behind a high net revenue retention rate. Others exist, like cross-selling add-on modules or gradually raising prices for existing customers, but in every case the common thread is the same: the typical customer spends more the following year without the vendor needing to start from zero to convince them, unlike acquiring an entirely new customer, which requires a whole costly sales cycle from scratch.

Why my quality filter does not calculate this number directly

If this number is so telling, a fair question arises: why does my analysis method, which judges a company's quality against 10 objective financial criteria, not include it directly in its scorecard? The answer comes down to a practical constraint rather than a lack of interest in what this number measures. Net revenue retention follows no universal accounting standard: every company chooses its own definition of the customer scope to track, its own measurement window, and crucially, nothing requires a public company to disclose it at all. Many software vendors report it voluntarily because it flatters them, but I cannot build a reliable, comparable criterion across companies from a figure that half of my analysis universe does not even make public, and that the other half could calculate differently without my being able to verify it from the outside.

What I do instead is observe the consequences of this mechanism in numbers I can verify uniformly across any public company: multi-year revenue growth trajectory (sustained customer expansion shows up as growth that holds up better than a company relying solely on new customers), free cash flow margin evolution (the pricing power that often accompanies strong retention shows up as margins expanding rather than compressing), and return on invested capital. Taken together, these three angles capture much of what net revenue retention reveals directly, without depending on a figure the company itself chooses whether or not to disclose.

The lesson goes beyond just software vendors: whenever you look at a subscription business, never settle for the headline revenue growth rate printed at the top of an earnings release. Always ask where that growth comes from, and specifically whether the customers who were already there last year are spending more or less than before. That is a harder question to answer precisely than to ask, given data is not always available, but it is exactly the kind of question that separates a solid investment thesis from a superficial read of a single number at the top of a press release.

FAQ

What is net revenue retention?

It is the percentage of revenue generated by a specific group of existing customers that is still there a year later, counting expansion (customers spending more) minus contraction and churn (customers spending less or leaving), excluding any new customers.

Why is a rate above 100% such a strong signal?

Because it means existing customers alone are already spending more than the year before: the company grows even without winning a single new customer, a sign of pricing power and a product becoming more essential over time.

Why do software companies above 120% net retention get valued so highly?

Market data shows a valuation gap of roughly 9.3 times revenue above 120% net retention versus roughly 3.1 times below 100%, a threefold factor that reflects investor confidence in the durability of growth when it comes from existing customers.

Why does my site not calculate this number for every stock?

Because there is no universal accounting standard governing it, and not every company discloses it. I would rather observe its consequences in data verifiable across every company (revenue trajectory, margins, return on capital) than rely on a figure each company chooses whether to disclose.

About the author

Written by Lubin Danilo, founder of Lubin Investment. A self-taught individual investor, I find fundamental analysis fascinating, and it has delivered excellent results. For three years now, my performance has beaten the S&P 500. But analyzing every stock took too much time: sites with incomplete data, calculation methods and criteria never aligned with mine. And spotting the best stocks was just as time-consuming, even with my own well-defined checklist. So I put my software development background to work to build this software, base my investment strategy on its results, and share it with people who share the same passion as me. It judges a company's quality and its price separately, using criteria drawn from the financial literature (Warren Buffett, Michael Mauboussin, Aswath Damodaran).