Software-as-a-Service (SaaS) is the dominant business model in modern enterprise software: customers rent access to cloud-hosted applications on recurring subscriptions rather than buying a perpetual license. For an investor, this is a playbook — a distinct lens for reading software companies because their economics, metrics, and valuation logic differ sharply from most of the market. The core tension that runs through the whole sector: a SaaS firm pays the full cost of winning and serving a customer up front, in cash, but recovers it slowly over years of subscription revenue — and the entire investment case rests on the durability of that future stream. That makes software simultaneously one of the highest-quality business models ever (high gross margin, recurring revenue, operating leverage, expansion-driven compounding) and one of the most fragile to mis-underwrite, because almost everything that matters is a forward estimate, much of it reported in non-GAAP metrics with no standard definition.

What defines the model

Three structural features separate SaaS from most equities and justify a dedicated playbook:

• Recurring revenue. Revenue arrives as subscriptions that renew, producing high visibility and low volatility versus transactional businesses. This is why the sector is described in run-rate terms (ARR) rather than just trailing revenue.
• High gross margin and operating leverage. Public SaaS companies typically run ~70–75% gross margins (Software Equity Group; SaaSDB), and infrastructure cost grows sublinearly with revenue, so margins tend to expand at scale. Software Equity Group reports companies above 70% gross margin traded at a median ~7.2x EV/revenue versus ~3.5x for those below 60% — margin is a direct valuation input.
• Expansion economics. A healthy SaaS base can grow revenue without adding customers (net revenue retention >100%), creating a compounding flywheel that transactional businesses lack.

The trade-off is heavy up-front cash burn and dependence on estimates of churn and retention that may not hold — the recurring-revenue strength and the underwriting fragility are two faces of the same model.

The metrics that run the sector — map of sub-topics

This is a metrics-driven sector, and the child nodes cover the three metric families analysts actually use. Read those for depth; this overview only points to them:

• ARR & Net Revenue Retention (child node) — the level (annualized recurring run-rate) and the quality-of-growth rate (how an existing cohort grows once upsells net against churn). NRR >100% signals a compounding base; the gap between NRR and gross retention (GRR) exposes churn hidden under expansion. Both are unaudited non-GAAP figures with no standard definition.
• Rule of 40 (child node) — the back-of-envelope health check: revenue-growth % + profit-margin % ≥ 40, capturing the growth-vs-profitability trade-off in one number. Widely cited but with an arbitrary threshold, no standard margin definition, and contested equal-weighting (Bessemer's "Rule of X" argues growth deserves a multiplier).
• CAC, LTV & Churn (child node) — the unit economics: what a customer costs to acquire, the gross profit they throw off, and the churn rate that sets their lifetime. LTV:CAC and CAC payback period gauge whether the acquisition engine actually makes money. The headline caveat is the constant-churn assumption baked into LTV.

A practical "Big 5" framing (SaaS Capital / Consero) ties them together: ARR growth, net retention, gross retention, recurring gross margin, and CAC/payback — ARR growth driving the most valuation impact, retention validating product-market fit.

How the sector is valued

Because many SaaS firms are deliberately unprofitable (reinvesting in growth), EV/Revenue (or EV/ARR) is the dominant multiple, not P/E or EV/EBITDA. The multiple is driven by growth rate, net retention, gross margin, and — increasingly — AI defensibility. SaaS Capital and others document that Rule-of-40 outperformers and high-NRR names command premium multiples; analysts cite roughly a couple of turns of additional EV/Revenue per ~10 points of Rule-of-40 score (directional, source-dependent). As businesses mature, valuation attention shifts from pure growth toward FCF margin and profitability.

When the playbook matters — and its biggest risk: duration

This lens matters for any cloud/subscription software name; it is less applicable to hardware, semiconductors, transactional/marketplace tech, or legacy perpetual-license software (where the recurring-revenue and retention metrics distort or don't apply). The single most important sector-wide risk an investor must hold is that software is a long-duration asset: most of its value sits in cash flows far in the future, so it is acutely sensitive to the discount rate. When the Fed raised rates in 2022, growth software sold off hardest — the Aventis SaaS Index fell more than 55% from peak — precisely because raising the discount rate cheapens distant cash flows. This duration sensitivity is a defining macro behavior of the sector, not an idiosyncratic event.

Adoption, debate & evidence

The SaaS metric vocabulary (ARR, NRR/NDR, Rule of 40, LTV/CAC) is near-universal in S-1s, earnings releases, and VC diligence. But two honest caveats define the sector's evidence base:

1. The metrics are non-GAAP and gameable. ARR has no regulator-set definition; NRR depends on cohort window and netting choices; Rule-of-40 inputs (EBITDA vs FCF vs net margin, ARR vs GAAP growth) are not standardized. Cross-company figures are rarely strictly comparable, and the SEC keeps non-GAAP measures among its top comment areas. The child nodes detail these traps. 2. The "AI disruption" thesis is the live, genuinely contested debate. Through 2025–early 2026, a sharp bear case emerged that AI agents erode SaaS moats (data lock-in, seat-based pricing, workflow habit) and could compress revenue-per-customer — and the market repriced it violently: the S&P 500 Software & Services index lost roughly $2 trillion in market value from its October 2025 peak through early 2026 ("SaaSpocalypse"), with pure-SaaS median EV/Revenue bottoming around ~3.1–3.4x — decade lows — and, for the first time, software trading at a discount to the S&P 500 (SaaStr; Aventis). The counter-case argues durable systems-of-record survive as orchestration layers under AI, that disruption is uneven (CRM/PM most exposed), and that early agent rollouts (e.g. Klarna's, which partly walked back full automation) overstated displacement. This is unresolved — treat any confident verdict, bull or bear, with suspicion.

Strengths & limitations

Strengths: the playbook captures real economic quality GAAP smears out — recurring revenue, expansion-driven compounding, and operating leverage — and gives a consistent, comparable framework for screening and valuing software names.

Limitations / #1 misuse: treating these metrics as audited facts or pass/fail verdicts. Every headline number embeds estimates and definitional choices; a strong NRR can mask weak gross retention, a single-quarter Rule-of-40 reading is near noise, and a computed LTV swings wildly with small churn errors. And the whole sector's valuation is hostage to the discount rate and, now, to an unsettled AI thesis — fundamentals can be excellent while the multiple halves on macro or narrative.

Sources

• Software Equity Group — gross margin and SaaS valuation: https://softwareequity.com/blog/gross-margin-saas/ ; Rule of 40: https://softwareequity.com/blog/rule-of-40/
• SaaSDB — SaaS gross margin benchmarks (~70–75%): https://saasdb.app/learn/financials/gross-margin/
• Aventis Advisors — SaaS valuation multiples 2015–2026 (index drawdown, EV/revenue): https://aventis-advisors.com/saas-valuation-multiples/
• SaaStr — "The SaaS Rout of 2026 / software trades at a discount to the S&P 500": https://www.saastr.com/the-saas-rout-of-2026-is-even-worse-than-you-think-for-the-first-time-ever-software-now-trades-at-a-discount-to-the-sp-500/
• Consero — SaaS / recurring-revenue KPIs ("Big 5"): https://conseroglobal.com/resources/the-guide-to-key-metrics-for-saas-and-recurring-revenue-models/
• Sapphire Ventures — 2026 Software x AI inflection point: https://sapphireventures.com/blog/2026-softwares-ai-inflection-point/
• Twig — Klarna AI customer-support case (partial walk-back): https://www.twig.so/blog/klarna-ai-customer-support-efficiency
• Child nodes (this section): ARR & Net Revenue Retention; Rule of 40; CAC, LTV & Churn

Disputes flagged: (1) SaaS headline metrics are non-GAAP, unstandardized, and not strictly comparable across companies. (2) Valuation multiples and benchmarks vary materially by source/date and were in steep, ongoing decline through early 2026 — all figures are directional snapshots. (3) The AI-disruption thesis for SaaS is actively contested and unresolved; no efficacy verdict is asserted here.