- name:
- analyzing-saas-business-metrics
- language:
- en
- description:
- Deconstructs SaaS operating metrics including ARR, NRR, gross retention, magic number, rule of 40, and cohort economics. Use when analyzing SaaS businesses, benchmarking software metrics, or evaluating subscription model health.
- author:
- casemark
Analyzing Saas Business Metrics
Deconstructs SaaS operating metrics including ARR, NRR, gross retention, magic number, rule of 40, and cohort economics.
When To Use
- Evaluating a SaaS company for growth equity or late-stage investment
- Benchmarking a portfolio company's operating metrics against peer cohorts
- Assessing subscription model health during due diligence or quarterly reviews
- Analyzing expansion revenue dynamics and churn trends for board-level reporting
- Comparing "efficient growth" profiles across competing deal opportunities
Inputs To Gather
- Revenue data: Monthly or quarterly MRR/ARR schedules covering at least 12–24 months; breakdown by new, expansion, contraction, and churned revenue
- Customer data: Beginning-of-period and end-of-period customer counts by cohort (sign-up quarter or ACV tier); logo churn counts
- Expense data: Fully loaded S&M spend by quarter (including headcount costs); COGS broken out to show hosting, support, and CS costs separately
- Cohort schedules: Revenue by vintage cohort showing retention and expansion over time (ideally 8+ quarterly cohorts)
- Billing & contract details: Contract term mix (monthly vs. annual vs. multi-year), average ACV, and gross-to-net revenue adjustments
- Context: Industry vertical, target customer segment (SMB/mid-market/enterprise), and pricing model (seat-based, usage-based, platform fee)
Workflow
-
Validate the ARR bridge
- Reconstruct ARR from Beginning ARR + New + Expansion − Contraction − Churn = Ending ARR
- Confirm each component ties to underlying MRR schedules; flag any reconciliation gaps with [VERIFY]
- Distinguish between committed ARR and run-rate ARR if usage-based revenue is material
-
Calculate core retention metrics
- Gross Dollar Retention (GDR): (Beginning ARR − Contraction − Churn) / Beginning ARR. Healthy benchmark: >90% for enterprise, >80% for SMB [VERIFY against current market benchmarks]
- Net Dollar Retention (NDR/NRR): (Beginning ARR + Expansion − Contraction − Churn) / Beginning ARR. Best-in-class: >120% enterprise, >110% mid-market
- Logo retention: 1 − (churned logos / beginning logos). Separate from dollar retention to isolate unit economics from mix effects
- Note whether retention is calculated on a trailing-12-month basis or quarterly annualized — results differ materially
-
Assess growth efficiency
- Magic Number: Net New ARR (current quarter) / S&M Spend (prior quarter). Targets: >0.75 indicates efficient spend; >1.0 is strong; <0.5 signals concern
- CAC Payback: Fully loaded S&M per new customer / (new ARR per customer × gross margin). Express in months; <18 months is generally attractive for enterprise SaaS
- LTV/CAC: (ARR per customer × gross margin) / (1 − GDR) / CAC. Minimum threshold typically >3x
- Separate new-logo acquisition efficiency from expansion-driven growth — blended metrics can mask deteriorating new-business economics
-
Evaluate profitability and the Rule of 40
- Rule of 40 = YoY ARR growth rate (%) + FCF margin (%) or EBITDA margin (%). Score ≥40 is the benchmark; identify which component is driving the result
- Break out gross margin (target >70% for pure SaaS, >60% if managed services are included [VERIFY]). Scrutinize hosting costs, capitalized development, and professional services pass-throughs
- Examine operating leverage: is S&M as a % of revenue declining as revenue scales? Is G&A staying flat or growing?
-
Analyze cohort economics
- Build a cohort waterfall: for each vintage, show cumulative revenue as a multiple of first-period revenue at 4, 8, 12, and 16+ quarters
- Identify whether newer cohorts retain and expand at the same rate as older ones — declining cohort quality is a red flag
- Calculate cohort payback: quarters until cumulative gross profit from a cohort exceeds the S&M cost to acquire it
- Flag if expansion is concentrated in a small number of large accounts vs. broad-based
-
Benchmark and contextualize
- Compare all key metrics against relevant public SaaS comps and private benchmarks (e.g., Bessemer Cloud Index, ICONIQ Growth reports, KeyBanc SaaS survey) [VERIFY data sources are current]
- Adjust for company stage (e.g., $10M ARR vs. $100M ARR expectations differ), end-market, and contract structure
- Highlight where the company is top-quartile, median, or below-median on each metric
Output
Produce a structured SaaS Metrics Analysis containing:
- Executive summary: 3–5 sentence assessment of subscription model health and growth quality
- ARR bridge table: Quarterly ARR waterfall with all components for the trailing 8+ quarters
- Retention dashboard: GDR, NDR, and logo retention trended over time with commentary on drivers
- Efficiency scorecard: Magic number, CAC payback, LTV/CAC, and Rule of 40 — each with a benchmark comparison
- Cohort matrix: Vintage cohort retention/expansion grid with visual highlighting of trends
- Key findings: Bulleted list of 5–10 specific observations (strengths, risks, and areas requiring further diligence)
- Open items: Any metrics that could not be calculated due to missing data, marked [VERIFY]
Quality Checks
- ARR bridge reconciles to within 1% of reported ending ARR; any variance is explained
- Retention metrics are calculated on a consistent basis (same period definition) throughout
- Magic number uses prior-quarter S&M spend (not same-quarter) to reflect the sales cycle lag
- Gross margin excludes capitalized software development costs unless explicitly noted
- Cohort analysis covers enough vintages (minimum 4) to identify trend vs. noise
- All benchmark comparisons cite the source and vintage of the benchmark data
- No metric is presented without context — raw numbers always accompanied by what "good" looks like for the company's stage and segment
- Assumptions about annualization, seasonality adjustments, or pro-forma treatments are stated explicitly