The 7 KPIs Every SaaS CFO Should Own — and 3 They Should Stop Tracking

When capital was cheap, growth at any cost was a strategy and the metrics followed suit: bigger top-of-funnel numbers, bigger user counts, bigger growth percentages. That era ended, but its dashboards did not. We still open board decks where the first three slides celebrate registered users and MQL volume while net revenue retention hides in an appendix. The metrics below are the ones we help finance teams compute reliably — and the ones we help them retire.

The Seven to Own

• Net Revenue Retention (NRR) — revenue from your existing customer base twelve months ago versus that same cohort today, including expansion, contraction, and churn. Above 110% means the installed base grows on its own; below 100% means you are refilling a leaking bucket. NRR is the single best predictor of efficient growth, which is why it now leads serious investor conversations.
• Gross Revenue Retention (GRR) — the same calculation with expansion stripped out. It answers a colder question: how much revenue do we simply keep? A company with 115% NRR and 80% GRR is masking heavy churn with a few expanding whales. Owning both numbers keeps the story honest.
• CAC Payback Period — fully loaded sales and marketing cost to acquire a customer, divided by the gross-margin-adjusted monthly recurring revenue they generate. Under 18 months is healthy for mid-market; under 12 is excellent. It replaces the gameable LTV:CAC ratio because it requires no speculative lifetime assumptions — just how fast cash comes back.
• Rule of 40 — revenue growth rate plus free-cash-flow margin should exceed 40. It is the one-line answer to whether you are buying growth with burn or earning it. A company growing 60% while burning 30% of revenue scores 30 — and that arithmetic ends the "but look at our growth" defence.
• Gross Margin, cloud-cost-aware — true SaaS gross margin nets out hosting, third-party API costs, and the human cost of support and customer success. Many companies reporting 80% margins are closer to 65% once cloud spend is correctly allocated per customer and per feature. AI features make this worse: inference costs can quietly turn a profitable tier into a subsidised one.
• Net Burn Multiple — net cash burned divided by net new ARR added in the same period. Below 1x is elite; above 2x means every dollar of growth costs more than two dollars of runway. It is the efficiency metric that survives any pricing model or market cycle.
• Free-Cash-Flow Margin — the bridge from SaaS metrics to actual finance. Deferred revenue timing, annual prepays, and capitalised development costs all distort EBITDA; FCF margin cuts through them. Paired with growth in Rule of 40, it tells the only story that matters at exit.

The Three to Retire

• Raw MRR growth % without cohort context — "MRR grew 8% this month" is unfalsifiable theatre. Was it new logos, expansion, a price increase, or one enterprise deal that distorts everything? Without cohort decomposition, the number cannot inform a single decision. Replace it with an MRR bridge: new, expansion, contraction, churn — by cohort.
• Total registered users — a count that includes everyone who ever clicked "sign up," including the trial accounts from 2019. It only goes up, which is precisely why it measures nothing. Active accounts against a strict activity definition, or seats under paid contract, are the honest substitutes.
• Gross MQL volume — when marketing is targeted on MQL count, marketing manufactures MQLs; the metric corrupts the behaviour it measures. Pipeline dollars created and pipeline-to-revenue conversion by source cost more effort to compute and repay it in every forecasting conversation.

Why the Hard Part Is the Data, Not the Definitions

Every CFO can recite these definitions. Far fewer can compute them without a week of spreadsheet archaeology, and that gap is architectural. NRR requires a clean monthly snapshot of revenue per customer with consistent entity resolution — so that "Acme Corp," "Acme Corporation," and the subsidiary that signed its own contract roll up to one customer. CAC payback requires marketing spend, sales compensation, and revenue to share common time grains and attribution logic across systems that were never designed to agree. Cloud-cost-aware gross margin requires tagging discipline in your infrastructure and a cost-allocation model that maps spend to customers and features.

This is exactly the work we do at One Big Table: building the unified data layer where billing, CRM, product telemetry, and cloud cost data resolve into one governed model, so that every one of these KPIs is computed one way, from one certified source, on a refresh schedule finance can trust. When the metric definitions live in a semantic layer instead of in seventeen competing spreadsheets, the quarterly close stops being an act of reconciliation and starts being an act of analysis. Our methodology typically starts with the metric definitions and works backwards to the pipelines — not the other way round.

A practical first step: audit your current board deck against the list above. Every metric that appears in the deck but not in the list needs a defender; every metric in the list but not in the deck needs a pipeline. If you want a structured read on where your reporting stands, our Data Maturity Index takes fifteen minutes and benchmarks exactly this.

Follow OBT on LinkedIn