Gross Margin for SaaS: What It Is, How to Calculate It, and Why It Determines Valuation
Gross margin is the foundation of every SaaS valuation model. A 10-point improvement can increase your company's value by 30–50% at the same ARR. Here's everything you need to know.
What Gross Margin Actually Measures
Gross margin is the percentage of revenue remaining after subtracting the direct cost of delivering your product. For SaaS: what percentage of each subscription dollar do you keep after paying for infrastructure, APIs, payment processing, and support required to deliver the product? At 75% gross margin, you keep $0.75 of every $1 in revenue to cover operating expenses, R&D, sales, and profit. Calculate your exact gross margin using our free gross margin calculator.
What Counts as COGS for SaaS
COGS includes only costs directly tied to delivering the product:
- Cloud infrastructure (AWS, GCP, Azure)
- Payment processing fees (typically 2–3% of revenue)
- Third-party API costs (AI models, mapping, SMS)
- Customer support costs for serving existing customers
COGS does not include product engineering, R&D, sales, marketing, or G&A. Including them artificially suppresses reported gross margin.
Gross Margin Benchmarks
- Pure SaaS: 75–85% standard; 85–90%+ best-in-class
- SaaS + professional services: 60–70% typical
- AI-native SaaS: 55–70% due to inference costs
- Hardware-bundled SaaS: 50–65%
Most Series A/B investors expect gross margins above 70%. Below that, you'll face cost structure questions in every diligence conversation.
Why Gross Margin Drives Valuation
A 10-percentage-point gross margin improvement increases valuation by 30–50% at the same ARR. The mechanism: higher gross margin means more of each incremental revenue dollar flows to operating income as the business scales. A 80% GM business at $20M ARR generates $16M in gross profit. A 55% GM business generates only $11M — a $5M difference that compresses every profitability metric downstream.
How to Improve Gross Margin
Infrastructure optimisation typically yields 20–40% cost reduction: right-sized instances, reserved capacity, reduced egress, aggressive caching. Payment processing — renegotiating rates at volume thresholds, moving enterprise to ACH — recovers 0.5–1% of revenue. AI inference costs require active management: cache repeated queries, use smaller models for routine tasks, negotiate volume pricing. Support cost reduction through self-service tooling and in-app guides reduces per-customer COGS directly.