Back to Glossary
MetricsG

Gross Margin

Definition

Gross margin is the percentage of revenue remaining after subtracting the cost of goods sold (COGS). The formula is: (Revenue - COGS) / Revenue x 100. For a SaaS company, COGS typically includes hosting and infrastructure costs, third-party software licenses embedded in the product, customer support and onboarding staff costs, and payment processing fees.

A company with $10M in annual revenue and $2.5M in COGS has a gross margin of 75%. That means $7.5M is available to cover operating expenses (engineering, sales, marketing, G&A) and generate profit. Gross margin is the first line of defense in a company's financial model -- if it's too low, no amount of revenue growth can make the business sustainable.

Why It Matters for Product Managers

PMs who ignore gross margin make expensive mistakes. Every architectural decision, third-party integration, and support model choice has a COGS implication. Building a feature that calls an expensive external API on every user action, requiring manual data review before delivering results, or offering unlimited storage -- these decisions erode margin in ways that aren't visible in product metrics but show up painfully in financial reviews.

The AI/ML wave has made this especially relevant. Products that use large language models face per-query inference costs that can be 10-100x more expensive than traditional API calls. A PM at an AI-powered product needs to understand that a feature generating 1M API calls/day at $0.01/call costs $10K/day -- $3.6M/year. If that feature generates $5M in revenue, the gross margin is only 28%, far below the SaaS benchmark. Optimizing prompt length, caching responses, and choosing the right model size are product decisions with direct margin impact.

Gross margin also determines how much a company can invest in growth. High-margin companies (like Atlassian at 83%) can afford to spend more on R&D and go-to-market while remaining profitable. Lower-margin companies face hard tradeoffs between growth investment and profitability.

How It Works in Practice

  • Know your COGS line items -- Ask your finance team for a breakdown of COGS by category. Common SaaS COGS: cloud hosting (AWS/GCP/Azure), third-party APIs, payment processing, customer support labor, and data infrastructure. Understand which costs are fixed vs. variable.
  • Calculate margin by feature or product line -- Aggregate gross margin hides important variation. A core product feature at 90% margin can subsidize an AI feature at 40% margin, but you should know where each stands. Work with finance and engineering to allocate infrastructure costs by feature.
  • Model margin impact of new features -- Before greenlighting a feature, estimate its COGS impact. "This feature requires 10 GPU-hours per user per month at $2/hour" is the kind of estimate PMs should produce during planning. Compare the incremental revenue the feature drives against its incremental COGS.
  • Optimize existing features -- Partner with engineering to reduce COGS on shipped features: implement caching layers, optimize database queries, negotiate better vendor rates, move to reserved instances, or choose cheaper model sizes for AI features where quality isn't affected.
  • Track margin trends -- Monitor gross margin monthly. If it's declining as you scale, you have a structural cost problem. If it's improving, your unit economics are getting healthier. Share margin trends with the product team so everyone understands the financial impact of their decisions.
  • Common Pitfalls

  • Treating infrastructure as "someone else's problem." In a product-led company, the PM owns the user experience end to end -- including the cost of delivering it. If your feature choice drives $500K/year in hosting costs, that's a product decision with financial consequences.
  • Confusing gross margin with net margin. Gross margin only subtracts COGS. Net margin subtracts everything (R&D, sales, marketing, G&A). A company can have excellent gross margins and still lose money if operating expenses are too high. PMs affect gross margin directly; they affect net margin indirectly through team size and tool choices.
  • Ignoring margin when celebrating revenue growth. A new enterprise contract worth $500K/year at 30% margin is less valuable than a $300K/year self-serve cohort at 85% margin. Revenue without margin context is misleading.
  • Over-optimizing margin at the expense of product quality. Reducing support staff, choosing the cheapest hosting tier, or using a smaller AI model to save costs can degrade the experience enough to increase churn. Margin optimization should never sacrifice the core value proposition.
  • Understanding burn rate alongside gross margin gives you the full financial picture -- burn rate tells you how fast you're spending cash, while gross margin tells you how efficiently you're generating it. ARR/MRR is the revenue number that gross margin is calculated against, so tracking both together shows whether revenue growth is healthy or masking deteriorating economics. Gross margin directly flows into LTV calculations since higher-margin customers generate more lifetime value per revenue dollar.

    Frequently Asked Questions

    What is a good gross margin for a SaaS company?+
    Best-in-class SaaS companies operate at 75-85% gross margins. Below 70% raises concerns with investors because it limits how much you can spend on sales, marketing, and R&D while maintaining a path to profitability. Companies with heavy infrastructure costs (like AI/ML products with GPU compute) or high-touch services often run lower margins, typically 50-65%.
    Why should product managers care about gross margin?+
    Every product decision affects gross margin. Choosing to host on GPUs instead of CPUs, building features that require manual data processing, adding human-in-the-loop workflows, or offering white-glove onboarding all increase COGS and reduce margin. A PM who ships a feature that drives $1M in revenue but costs $800K to deliver has created a 20% margin feature that drags down company economics.

    Explore More PM Terms

    Browse our complete glossary of 100+ product management terms.