SaaS Metrics: MRR, Churn, NRR and What Investors Look At

Forty minutes with an investor, and the first three questions are always about SaaS metrics: what's your MRR, how fast is it growing, and how much of it do you keep. Founders who answer from memory get the rest of the meeting to talk about product. Founders who open a spreadsheet on the spot spend the whole meeting explaining themselves.
The trouble is that the topic has turned into alphabet soup: MRR, ARR, NRR, GRR, CAC, LTV, payback, Rule of 40. Not all of it matters all the time, and memorizing acronyms without understanding the mechanics is worse than not memorizing them at all. This pocket guide covers the formulas worth knowing by heart, with worked examples and market benchmarks that come with named sources.
SaaS metrics are the indicators that measure recurring revenue and efficiency in a subscription business. The essential ones: MRR (monthly recurring revenue), churn (lost customers or revenue), NRR (net revenue retention), CAC (customer acquisition cost), LTV (customer lifetime value) and CAC payback. Together, they show whether growth is healthy or subsidized.
MRR and ARR: the starting point
MRR (Monthly Recurring Revenue) is the sum of all active subscriptions normalized to one month. A customer paying $1,200 per year counts as $100 of MRR. ARR is simply MRR multiplied by 12, favored by companies selling annual contracts.
What stays out: setup fees, one-off services, consulting, any revenue that doesn't repeat. Mixing one-time revenue into recurring revenue is one of the most common mistakes in SaaS pitches, and an experienced investor spots it by the second question.
More important than the headline number is the breakdown of the month's movement:
- New MRR: revenue from customers who signed up this month.
- Expansion MRR: upgrades and upsells from existing customers.
- Contraction MRR: downgrades from customers who stayed but pay less.
- Churned MRR: revenue from customers who cancelled.
A round-numbers example: a company starts the month at $100k MRR, closes $12k in new customers, expands $5k within the base, takes $3k in contraction and loses $6k to cancellations.
MRR grew 8% for the month, which looks great. But take the new customers out of the equation and the existing base went from $100k to $96k. Gross growth is hiding weak retention, and that is exactly the trick NRR exposes, as we'll see below.
Churn: how much you lose every month
Customer churn is the percentage of accounts that cancel in a period: customers lost divided by customers at the start of the month. Revenue churn uses cancelled MRR instead of account counts, and it often tells a different story when ticket sizes vary widely. Losing ten $99 customers hurts less than losing one $5k account.
The detail many people underestimate: monthly churn compounds. Churning 3% per month means losing roughly 31% of your base in a year. At 5% per month, the annual loss approaches 46%. It's like filling a pool with the drain open: the acquisition tap has to run faster and faster just to keep the level steady.
In practice, tolerable churn depends on who you sell to. Low-priced self-serve products for small businesses live with far higher monthly churn than enterprise annual contracts, where the expectation is close to zero. Comparing the two without that context produces either panic or complacency, both unwarranted. The full formulas and reduction tactics are in our churn rate guide.
NRR: the first metric investors look at
NRR (Net Revenue Retention) answers a simple question: if you sold to zero new customers, would your revenue grow or shrink? The formula only considers the base that already existed at the start of the period:
NRR = (starting MRR + expansion − contraction − churn) ÷ starting MRR
In the waterfall example: (100 + 5 − 3 − 6) ÷ 100 = 96%. Below 100%, the base shrinks on its own and all growth depends on new sales. Above 100%, the company grows even standing still, because expansion from current customers more than offsets the losses.
To calibrate expectations with real data: the 2026 Aleph and Benchmarkit benchmark, covering 342 SaaS and AI companies, puts median NRR at 102%, rising to 108% for usage-based pricing and dropping to 98% for seat-based models. In other words: 110% NRR already puts a company in elite territory, and 95% is nothing to be ashamed of for low-ticket products.
It's also worth knowing GRR (Gross Revenue Retention), which is NRR without expansion. It never exceeds 100% and measures only how much revenue you hold on to. High NRR with low GRR means a few heavily expanding customers are masking a bleed across the base.
CAC, LTV and payback: the cost of growing
CAC is total marketing and sales spend divided by customers won in the period. LTV is the margin a typical customer generates before cancelling. The full formulas are in our CAC and LTV guides; what matters here is how they combine.
Example: CAC of $3,000, a $500-per-month plan and 80% gross margin. Each customer returns $400 net per month, so payback is 7.5 months. With 2.5% monthly churn, average customer lifetime is 40 months and LTV comes to $16,000, an LTV/CAC ratio of 5.3.
The classic reference asks for LTV/CAC above 3 and payback under 12 months. But this formula breaks for young companies: with only a few months of history, estimated churn is a guess, and a guess raised to an exponent becomes fiction. Before roughly 18 months of data, trust payback, which is observable, over LTV, which is a projection.
Rule of 40: growth and profit in one number
The Rule of 40 adds annual revenue growth to profit margin (usually free cash flow or EBITDA margin). A result above 40 signals a healthy balance between growing and burning cash. A company growing 25% a year at an 18% margin scores 43 and passes; another growing 60% while burning 30% of margin scores 30 and raises a flag.
It's a yardstick for companies with some scale. A SaaS at $100k ARR should be worried about finding repeatable sales, not optimizing this sum. It enters the picture once the operation has a few million in annual revenue and cash starts driving decisions, as we cover in our guide to financial indicators.
Summary table: formulas and references
| Metric | Formula | Practical reference |
|---|---|---|
| MRR | Sum of active subscriptions normalized per month | Track the breakdown (new, expansion, contraction, churn) |
| ARR | MRR × 12 | Standard for annual contracts |
| Revenue churn | Lost MRR ÷ starting MRR | Near zero for enterprise; higher and tolerable for SMB |
| NRR | (Starting MRR + expansion − contraction − churn) ÷ starting MRR | Market median: 102% (Aleph/Benchmarkit, 2026) |
| GRR | NRR without expansion | Theoretical maximum of 100% |
| CAC payback | CAC ÷ (MRR per customer × gross margin) | Under 12 months |
| LTV/CAC | LTV ÷ CAC | Above 3, with 18+ months of history |
| Rule of 40 | Annual growth % + margin % | Above 40, for scaled operations |
Which metrics do investors look at first?
Capital for SaaS in Latin America remains active: according to an Acorn Capital study reported by Finsiders, the region's SaaS market should reach $46 billion by 2027, with Brazil concentrating about 48% of that volume. More money watching the sector also means a higher bar in due diligence.
In practice, a fund's reading order tends to be: NRR first, because it reveals product quality; then ARR growth, gross margin and CAC payback; finally, how much cash the company burns to sustain it all. Vanity numbers like downloads or registered users don't make the cut, and insisting on them in a pitch signals not knowing where the game is played.
Frequently asked questions
What's the difference between MRR and ARR?
MRR is recurring revenue normalized per month; ARR is the same value multiplied by 12. Monthly self-serve businesses usually talk MRR, while companies closing annual enterprise contracts report ARR. Two photos of the same thing at different scales.
What is a good NRR for SaaS?
The market median in the 2026 Aleph/Benchmarkit benchmark was 102%. Above 110% is elite territory, common in usage-based pricing models. Below 95%, growth gets too expensive, because every new dollar first has to replace what the base lost.
What's the difference between NRR and GRR?
NRR includes expansion revenue from current customers; GRR ignores expansion and measures only how much of the original revenue you retained. GRR never exceeds 100%. Watching both together prevents upsells concentrated in a few accounts from hiding widespread cancellations.
Should I track customer churn or revenue churn?
Both, because they diverge when ticket sizes vary. Customer churn shows product health across the broad base; revenue churn shows the real financial impact. If revenue churn is higher than customer churn, you're losing precisely the big accounts.
Does MRR replace accounting revenue?
No. MRR is a management metric: it ignores non-recurring revenue, taxes and accrual rules. It works for management and investor conversations, but official financials remain the accountant's job. Keeping both views separate avoids surprises in due diligence.
Measuring is easy; keeping it current is the game
None of these formulas is hard. What's hard is having the right numbers, up to date and consistent across CRM, billing and spreadsheets, when the question comes, whether from an investor or from your co-founder on a Tuesday night. Whoever calculates NRR once a quarter, the night before the board meeting, discovers problems three months late. With Sherlok, you connect your data sources and simply ask: "what was my NRR last quarter?" or "how much of my new MRR came from expansion?", and get the analysis ready, no spreadsheet required.
Your AI agents already know what to do in your account.
Turn them on free, no credit card.
- No credit card
- Read-only access
- Revoke it any time