Founders drown in dashboards. The truth is that a handful of metrics explain whether a SaaS business is healthy, and the rest are supporting detail. This guide covers the core metrics — what each one means, how to calculate it, a worked example, why it matters, and a rough benchmark — then shows which ones to prioritize at each stage and the mistakes that quietly distort all of them.

A quick framing before the formulas: almost every SaaS metric is built from two raw inputs — recurring revenue and customers. Get those two clean and consistent, and the rest are arithmetic. Get them dirty — mixing one-time fees into MRR, counting trial users as customers — and every downstream number lies to you.

Recurring revenue: MRR and ARR

Monthly Recurring Revenue (MRR) is the predictable subscription revenue you earn each month. Annual Recurring Revenue (ARR) is the annualized view, typically MRR multiplied by twelve. Early teams live in MRR because it reacts fast; later-stage teams report ARR for planning and fundraising. Keep one-time charges — setup fees, one-off services, hardware — out of both. Recurring means recurring.

Formula:

MRR = sum of all normalized monthly subscription values ARR = MRR × 12

Normalize before you sum. An annual plan billed at $1,200 up front is $100 of MRR, not $1,200 in the month it was paid. A quarterly plan at $300 is $100 of MRR. Mixing billing cadence into MRR is the single most common reason a founder's number disagrees with their accountant's.

Worked example. You have 80 customers on a $50/month plan and 20 on a $200/month plan:

MRR = (80 × $50) + (20 × $200) = $4,000 + $4,000 = $8,000 ARR = $8,000 × 12 = $96,000

MRR movement: where growth actually comes from

A single MRR number hides the story. The useful view breaks the month-over-month change into four components:

• New MRR — recurring revenue from customers acquired this month.
• Expansion MRR — additional recurring revenue from existing customers (upgrades, seats, add-ons).
• Contraction MRR — recurring revenue lost to downgrades by customers who stayed.
• Churned MRR — recurring revenue lost to customers who cancelled entirely.

The relationship that matters:

Net New MRR = New + Expansion − Contraction − Churned

Worked example. You start the month at $50,000 MRR. You add $6,000 New, $3,000 Expansion, lose $1,000 to Contraction and $2,000 to Churn:

Net New MRR = 6,000 + 3,000 − 1,000 − 2,000 = +$6,000

You end at $56,000. But notice: gross new revenue was $9,000, and $3,000 of it leaked straight back out. Two businesses can both report "+$6,000 net" while one is healthy and the other is sprinting on a treadmill. Watching the components is what separates durable growth from churn you are outrunning.

Churn: the leak you cannot ignore

Churn is the revenue or customers you lose in a period. There is no single churn number — there are three, and confusing them is a classic mistake.

Logo (customer) churn

The share of customers you lost.

Logo Churn (%) = (customers lost in period / customers at start) × 100

Example: start with 200 customers, lose 8 → 8 / 200 = 4% monthly logo churn.

Gross revenue churn

The share of recurring revenue you lost to cancellations and downgrades, before any offsetting expansion.

Gross Revenue Churn (%) = ((churned MRR + contraction MRR) / starting MRR) × 100

Example: start at $50,000 MRR, lose $2,000 to churn and $1,000 to contraction → 3,000 / 50,000 = 6%. Gross revenue churn can never be negative — it only measures losses.

Net Revenue Retention (NRR)

The sharpest single retention metric. It nets expansion against churn and contraction for a fixed cohort of existing customers — new logos are excluded.

NRR (%) = ((starting MRR + expansion − contraction − churned) / starting MRR) × 100

Example: a cohort starts at $50,000, expands by $4,000, contracts by $1,000, churns $2,000 → (50,000 + 4,000 − 1,000 − 2,000) / 50,000 = 102%. An NRR above 100% means your existing base grows on its own, before you sell to anyone new. That is the engine behind efficient scaling.

Why this matters more than logo churn: expansion from existing accounts can offset — or outrun — lost logos, producing net negative revenue churn even while a few customers leave. A product can lose 4% of its logos and still grow recurring revenue from the survivors. The reverse is the danger: a growing business with high gross churn is filling a leaky bucket, and acquisition only papers over the hole.

Rough benchmarks. Many B2B SaaS teams target monthly revenue churn below 2–3%; enterprise products with annual contracts often run under 1% monthly. SMB and self-serve products structurally churn higher because the switching cost is low. Always read churn against your segment, not against a single universal number.

LTV and CAC: is growth profitable?

Customer Lifetime Value (LTV) estimates the total value an account produces over its lifetime; Customer Acquisition Cost (CAC) is what you spend to win it. Neither means much alone — the relationship between them is what tells you whether growth is creating value or destroying it.

Calculating LTV

The common approximation starts from ARPU and churn:

LTV = ARPU / customer churn rate

The intuition: if you lose 4% of customers a month, the average customer stays 1 / 0.04 = 25 months, so LTV is 25 months of ARPU.

The gross-margin adjustment matters. Raw revenue is not value — serving the customer has a cost (hosting, support, payment fees). The more honest figure multiplies by gross margin:

LTV (margin-adjusted) = (ARPU × gross margin %) / customer churn rate

Worked example. ARPU is $100/month, monthly churn is 4%, gross margin is 80%:

LTV = ($100 × 0.80) / 0.04 = $80 / 0.04 = $2,000

Without the margin adjustment you would have claimed $2,500 — overstating each customer's worth by 25%. Always state whether an LTV is gross-margin-adjusted; comparing an unadjusted LTV to CAC flatters your economics.

Calculating CAC

CAC = total sales and marketing spend in a period / new customers acquired in that period

Worked example. You spend $20,000 on sales and marketing in a month and acquire 50 customers → CAC = 20,000 / 50 = $400.

Be honest about the numerator: include salaries, ad spend, tooling, and commissions, not just media cost. A "$400 CAC" that excludes the SDR's salary is fiction.

The LTV:CAC ratio

LTV : CAC ratio = LTV / CAC

Using the figures above: $2,000 / $400 = 5:1.

A commonly cited target is around 3:1. Below roughly 1:1 you lose money on every customer. Far above 3:1 — say 8:1 — is not always a trophy; it can signal you are underinvesting in growth and leaving the market to competitors. The ratio is a guide, not a scoreboard.

CAC payback period

The ratio ignores time. CAC payback fixes that — it is the number of months of gross profit needed to recover CAC:

CAC Payback (months) = CAC / (ARPU × gross margin %)

Worked example. CAC $400, ARPU $100, gross margin 80% → 400 / (100 × 0.80) = 400 / 80 = 5 months.

A frequently cited target is under 12 months; the strongest SMB models recover CAC in well under a year. Payback matters because it is a cash-flow metric: a healthy LTV:CAC with a 24-month payback still ties up cash you may not have, especially before you raise.

ARPU and ARPA: revenue per account

Average Revenue Per Account (ARPU) — sometimes written ARPA, "per account" — is recurring revenue divided by active accounts.

ARPU = MRR / number of active accounts

Example: $8,000 MRR / 100 accounts = $80 ARPU.

Rising ARPU usually means pricing, packaging, or upsell is working; flat ARPU with rising headcount means you are adding low-value accounts that cost the same to serve. ARPU also feeds the LTV formula, so a small ARPU gain compounds into a larger LTV gain. For account-based B2B, "per account" (a company) is more useful than per individual user; for self-serve, per user may be closer to reality. Pick the denominator that matches how you actually sell, and keep it consistent.

Composite health metrics

Once the core metrics are clean, three composites summarize overall health in a single line — useful for board updates.

• Rule of 40. growth rate (%) + profit margin (%). A combined score at or above 40 is a commonly cited sign of a healthy balance between growth and profitability. A company growing 60% while burning at a −15% margin scores 45 — fine. One growing 10% at a 10% margin scores 20 — under pressure to fix one side.
• Quick ratio. (new MRR + expansion MRR) / (churned MRR + contraction MRR). It asks: for every dollar of recurring revenue you lose, how many do you add? A quick ratio of 4 means you add four dollars for each one lost. Higher is healthier; near 1 means you are barely treading water.
• Magic number. (net new ARR in a quarter) / (prior quarter's sales and marketing spend). A rough read on sales efficiency — a result around or above 1 is often treated as a green light to invest more in go-to-market.

Cohort and retention curves

Averages hide whether your product is actually getting stickier. Cohort analysis groups customers by the month they joined and tracks each group's retained revenue over time. Plotting that produces a retention curve.

The shape is the whole point:

• A curve that keeps sliding toward zero means you have no durable base — every customer eventually leaves.
• A curve that flattens at, say, 70% means a loyal core sticks indefinitely; that flat tail is the foundation of compounding revenue.
• A curve that smiles — dips, then rises above 100% — means expansion within the cohort outruns churn. That is the signature of best-in-class NRR.

Reading recent cohorts against older ones answers a question no single-month metric can: is the product improving? If newer cohorts retain better than older ones at the same age, onboarding and product fit are working — even if this month's blended churn looks flat.

Which metrics to watch by stage

Not every metric earns your attention at every stage. Tracking all of them from day one is a way to feel busy while learning nothing.

Pre-product-market-fit

Watch retention / cohort curves and logo churn above all. Revenue is small and noisy; what you need to know is whether anyone keeps using the product. A flattening retention curve is the clearest signal of fit. MRR is worth tracking, but optimizing CAC before you have retention is premature — you would just be acquiring customers who leave.

Early growth (PMF reached, scaling acquisition)

Now MRR movement (new vs. expansion vs. churn), CAC, CAC payback, and LTV:CAC come into focus. You are spending to grow, so you must confirm the unit economics work before pouring fuel on them. Gross revenue churn should be trending down as the product matures.

Scaling

NRR, the Rule of 40, and magic number dominate the conversation, alongside the core revenue numbers. At scale the questions are efficiency questions: is the existing base self-expanding (NRR above 100%), and is each go-to-market dollar still productive? Investors at this stage benchmark you on these composites.

Common mistakes

• Mixing one-time revenue into MRR. Setup fees and services inflate MRR and make growth look better than it is. Keep them separate.
• Not normalizing billing cadence. Booking a $1,200 annual plan as $1,200 of MRR in one month creates a phantom spike and a phantom crash twelve months later.
• Comparing unadjusted LTV to CAC. Without the gross-margin adjustment, LTV is overstated and the ratio flatters you.
• Ignoring contraction MRR. Downgrades do not register as churn but subtract from net MRR just the same; a business can show flat net growth during active selling because contraction is eating the new revenue.
• Reporting only blended averages. A single churn or ARPU number hides wildly different segments. Cut by plan, cohort, and customer size.
• Treating involuntary churn as unavoidable. Failed renewals are recoverable revenue, not lost customers — and they rarely appear in cancellation data.
• Chasing a large dashboard. A metric you never act on is overhead. Watch a small set well.

How Waffo Pancake helps

Pancake's analytics dashboard reports MRR, churn rate, LTV, and ARPU in real time, alongside trend, distribution, and customer analysis — so the metrics above come from your live payment data rather than a spreadsheet you update by hand. Because Pancake is your Merchant of Record, that revenue view is already net of tax and reconciled to settlement, which means your MRR matches the money that actually lands — not a gross figure you later have to deflate.

Two practical notes. First, Pancake measures recurring revenue and customers; if you bill on usage, you will want an external metering tool to feed consumption data, since Pancake does not meter usage itself. Second, because failed renewals are handled on a retry (dunning) flow rather than dropped on the first decline, the churn you see is closer to genuine customer churn than to involuntary leakage — which keeps the numbers above honest.

This article is general information, not tax, legal, or financial advice. Specific metric definitions, calculation conventions, and operating decisions depend on your situation — consult a qualified professional.

Related articles

SaaS Metrics

How to Calculate & Benchmark Your SaaS Growth Rate (2026)

SaaS Metrics

Net Profit vs. Gross Profit: The Formula Every Founder Needs

SaaS Metrics

What Is Gross Revenue in SaaS? (And Why It Isn't Cash)