A SaaS company can add customers, grow recurring revenue, and still create less cash than expected. The usual reason is not a lack of demand. It is that the cost to serve each account is rising alongside revenue. To improve SaaS gross margins, operators need to treat infrastructure, support, third-party software, and pricing as connected operating decisions rather than separate line items.
Gross margin is one of the clearest tests of whether growth can scale. It shows how much subscription revenue remains after the direct costs required to deliver and support the product. That remaining margin pays for sales, marketing, product development, administration, and eventually profit.
Start With the Right Gross Margin Calculation
The standard calculation is straightforward:
Gross margin = (Revenue – Cost of Revenue) / Revenue
The hard part is defining cost of revenue consistently. For a SaaS business, this commonly includes cloud hosting and data storage, payment processing, third-party APIs used to deliver core functionality, customer support, implementation labor when it is directly tied to onboarding, and customer success resources dedicated to service delivery.
Accounting treatment varies by company and stage. A high-touch enterprise SaaS vendor may include a larger share of onboarding and support payroll in cost of revenue than a self-serve product. Neither approach is automatically wrong, but changing classifications to make the percentage look better does not improve the underlying economics.
Start by reviewing gross margin by product line, plan, customer segment, and cohort. A blended company-level number can hide a low-priced plan with expensive support requirements or a large customer with disproportionate data and API consumption. Those details determine where action will actually matter.
1. Reduce Infrastructure Waste Before Cutting Capacity
Cloud spend is often the first place leaders look, and for good reason. Compute, storage, data transfer, observability, and managed databases can expand quickly as usage grows. But broad cost cutting can damage reliability, slow product teams, and create expensive incidents.
The better approach is to connect infrastructure costs to customer behavior. Identify idle environments, overprovisioned instances, duplicate data pipelines, unnecessary log retention, and workloads that run continuously but only need scheduled processing. Then measure unit costs such as hosting cost per active user, per workflow, per gigabyte stored, or per transaction.
Commitment discounts and reserved capacity can lower costs materially for stable workloads. They are less appropriate for volatile demand or an early-stage product still changing its architecture. Use commitments where utilization is predictable, and retain flexible capacity where uncertainty is high.
2. Put Guardrails Around Expensive Usage
Usage-based features can increase expansion revenue, but they can also produce margin-negative customers. This is especially common with AI features, high-volume integrations, video processing, analytics, and data-heavy workflows.
Do not assume that a usage-based price model is profitable because it feels modern. Measure the fully loaded delivery cost of each unit, including the vendor API, compute, storage, and support burden. If a customer pays $200 per month but generates $350 in model inference charges, growth in that segment makes the business weaker.
Set included usage limits that align with normal customer value, then charge for overages or move heavy users into a higher plan. Rate limits, batch processing, caching, model routing, and feature controls can also protect margins without making the product less useful. The goal is not to punish adoption. It is to ensure the customers creating the most delivery cost are contributing fairly to it.
3. Reprice Plans Around Value and Cost to Serve
Pricing is one of the fastest levers to improve SaaS gross margins, yet many companies leave old plans in place because changing them feels risky. A price increase will not fix a product with weak retention, but underpricing a valuable product creates a permanent margin problem.
Review each plan against three questions: What outcome does the buyer receive? What does this account cost to support and deliver? Which features are creating the biggest variable costs? Entry-level plans should be easy to evaluate and adopt, but they should not include unlimited access to resources that are costly at scale.
Packaging can be more effective than a flat price increase. For example, reserve premium integrations, priority support, advanced reporting, higher data retention, or AI credits for plans that can sustain their delivery cost. This gives customers a clear upgrade path while protecting the economics of lower-tier subscriptions.
Grandfathered accounts deserve careful handling. Forcing every long-term customer onto a new price can create avoidable churn. A phased migration, renewal-based adjustment, or usage-based add-on often preserves the relationship and improves profitability over time.
4. Make Customer Support More Efficient, Not Less Available
Support costs are direct costs for many SaaS companies, particularly when customers need technical setup, integration help, or ongoing configuration assistance. The wrong response is to make support difficult to reach. That may reduce a near-term expense while increasing churn, refunds, and negative word of mouth.
Instead, find the causes of support demand. Repeated tickets about permissions, billing, imports, product errors, or integration failures are product and process signals. Better in-app guidance, clearer onboarding, improved error messages, and self-service account controls can reduce ticket volume while improving the customer experience.
Segment service levels by contract value and complexity. A strategic enterprise account may justify named success resources and live implementation. A low-ACV self-serve account generally needs a product-led onboarding path, searchable documentation, and responsive but scalable support. Offering the same service model to every account is rarely efficient.
5. Audit Third-Party Vendors as Cost of Revenue
The software stack that helps operate a SaaS business can quietly become part of the cost to deliver it. API providers, communication services, identity tools, analytics platforms, security products, support systems, and embedded services may charge by user, event, call, or data volume.
For each vendor, map the spend to the product capability or internal workflow it supports. If the customer never sees the capability, it may belong in operating expenses rather than cost of revenue, depending on your accounting policy. If customers directly depend on it, treat it as a delivery cost and manage it accordingly.
Look for duplicate tools, pricing tiers that no longer match actual usage, and vendor contracts with automatic overage charges. Consolidation can help, but it is not always the answer. Replacing a proven provider with a cheaper option can introduce security, reliability, or engineering costs that erase the savings. Evaluate total cost, not just the subscription invoice.
6. Improve Retention in the Segments With Healthy Economics
Gross margin and retention are connected. Retained customers spread acquisition and onboarding costs across more months of revenue, while expansion in high-margin accounts improves the overall revenue mix. But not every retained customer is equally valuable.
Identify segments with strong gross margin, low support needs, reliable payment behavior, and high renewal rates. These are often the customers worth targeting with customer success outreach, expansion offers, and product investment. Conversely, a segment with frequent onboarding labor, heavy infrastructure use, and low willingness to pay may need a new package or a more selective sales motion.
This is where operators should avoid chasing logo count alone. A smaller base of customers that renews, expands, and uses a product efficiently can be financially stronger than rapid growth from accounts that consume more than they pay for.
7. Build a Monthly Gross Margin Operating Review
Margin improvement becomes durable when it has an owner and a regular cadence. Finance can report the metric, but product, engineering, support, and go-to-market teams each influence it.
A practical monthly review should track revenue, cost of revenue, gross margin, infrastructure cost per usage unit, support cost per account, vendor spend tied to delivery, and margin by plan or customer segment. When a metric moves, assign a reason. Was it a customer mix shift, a cloud pricing change, an AI usage spike, a new implementation model, or an accounting reclassification?
Use the review to prioritize work based on economic impact and customer risk. The most attractive projects reduce a meaningful recurring cost without weakening product performance, security, or retention. A small engineering improvement that cuts API calls across thousands of accounts can be more valuable than a one-time vendor negotiation.
The practical target is not the highest possible gross margin on a spreadsheet. It is a margin profile that supports your product model, gives the company room to invest, and improves as revenue scales. When every team can see the cost to serve the customer, better decisions become much easier to make.
