About Cost Per Request Calculator

Calculate cost per request and ROI metrics for infrastructure. Part of the DevTools Surf developer suite. Browse more tools in the Calculators collection.

Use Cases

• Calculate the infrastructure cost per API call for pricing decisions
• Compare serverless vs containerized architecture costs at different traffic levels
• Identify the traffic threshold where moving to reserved instances becomes cheaper than on-demand
• Justify infrastructure spending to leadership with per-request unit economics

Tips

• Enter monthly infrastructure cost broken down by service (compute, storage, network) — the calculator attributes costs to requests based on your traffic volume
• Use the idle capacity slider to account for over-provisioning: a server running at 30% average utilization costs 3x more per request than one at 90%
• The ROI panel shows break-even analysis: at what request volume does each infrastructure tier become cost-effective

Fun Facts

• Amazon Lambda's pricing (per-request plus per-GB-second of compute) was introduced in 2014 and popularized 'serverless' economics — paying only for exact usage rather than reserved capacity. At low traffic, serverless can be 100x cheaper than dedicated servers.
• Google processes approximately 8.5 billion searches per day. At Google's reported infrastructure costs, a conservative estimate puts cost-per-search at under $0.001 — achieved through massive economies of scale and custom hardware (TPUs, Jupiter networking).
• The cost per gigabyte of cloud storage has fallen from approximately $0.15/GB/month in 2010 to under $0.02/GB/month in 2024 — a 7.5x reduction in 14 years, following a trajectory similar to Moore's Law for compute.

FAQ

What is a good cost per API request?

Highly variable by use case. Simple CRUD operations: $0.0001-0.001. AI inference endpoints: $0.001-0.10. Video processing: $0.01-1.00. The benchmark is whether your revenue-per-request exceeds your cost-per-request with sufficient margin.

How do I reduce cost per request?

The four levers: increase utilization (right-size instances), add caching (fewer origin requests), use reserved/spot instances (lower compute rates), and optimize code efficiency (fewer resources per request). Caching typically gives the highest ROI.

Should I use serverless or dedicated servers?

Serverless is cost-effective under ~1 million requests/month and for variable or unpredictable traffic. Dedicated servers or containers are more cost-effective at consistent high volume. The crossover point depends on your compute-to-memory ratio and idle time.