Home/Library/Reduce Step Functions and SQS Costs
How-to · AWS Cost Optimization · Updated June 2026

How to Reduce AWS Step Functions and SQS Costs

Orchestration and messaging bill by volume. Step Functions Standard charges $25 per million state transitions while Express charges $1 per million, and SQS is $0.40 per million requests. The wrong workflow type and chatty messaging quietly multiply the bill. Match the model to the workload.

Key takeaways

To reduce AWS Step Functions and SQS cost, choose the cheaper workflow type, cut state transitions, and batch messages. Step Functions Standard bills $25 per million state transitions while Express bills $1 per million requests, so moving high-volume short workflows to Express is the single biggest lever. SQS Standard is $0.40 per million requests versus $0.50 for FIFO, so batch up to 10 messages per call, use long polling to avoid empty receives, and use Standard queues where ordering is not required. Matching the model to the workload is part of the 31% average bill reduction across the 500+ environments we have optimized.

Last updated: June 2026 · Written by Morten Andersen and reviewed by Fredrik Filipsson · See, Cut, Lock, Run method

AWS Step Functions orchestrates multi-step workflows and Amazon SQS moves messages between services, and both are billed purely by volume. That makes them cheap at low scale and surprisingly expensive at high scale, especially when a Standard state machine runs a chatty workflow millions of times or a queue is polled empty all day. The fixes are architectural, not commitments: pick the right workflow type and stop paying per message for work you can batch. This guide is part of our complete guide to AWS cost optimization, the cluster pillar it links up to.

How much do AWS Step Functions and SQS cost?

AWS Step Functions Standard workflows cost $25 per million state transitions and Express workflows cost $1 per million requests plus duration, while Amazon SQS costs $0.40 per million requests for Standard queues. The Step Functions figures come from the AWS Step Functions pricing page: Standard bills $0.025 per 1,000 state transitions, which is $25 per million, with 4,000 transitions free per month, and each retry counts as a transition; Express bills $1 per million requests plus $0.06 per GB-hour of execution duration. The SQS figures come from the Amazon SQS pricing page: Standard queues are $0.40 per million requests, FIFO queues are $0.50 per million, the first 1 million requests are free each month, and each 64 KB chunk of a payload counts as a request. Verify current rates before modeling, since pricing changes by region.

How do I reduce Step Functions and SQS cost step by step?

Reduce cost by choosing the cheaper workflow type, trimming transitions, and batching messages. Work through these steps:

  1. Find your high-volume workflows and queues. Use AWS Cost Explorer and Amazon CloudWatch to identify which state machines and queues drive the most transitions and requests, since cost scales directly with volume.
  2. Move high-volume short workflows to Express. Convert high-throughput, short-duration state machines from Standard to Express, since Express is $1 per million requests versus $25 per million state transitions, a difference that dominates the bill at scale.
  3. Reduce state transitions per execution. Combine steps, remove unnecessary Pass and Wait states, and avoid aggressive retry loops, because each retry and each step bills as a state transition on Standard workflows.
  4. Batch SQS messages and use long polling. Send and receive up to 10 messages per API call and enable long polling so empty receives stop billing against the $0.40 per million request rate.
  5. Use Standard queues where ordering is not required. Choose Standard SQS at $0.40 per million over FIFO at $0.50 unless strict ordering or exactly-once processing is genuinely needed.

This is the same waste-elimination work in our AWS cost optimization service. The encryption-and-secrets counterpart, where keys and secrets each carry a small recurring charge, is covered in the sibling guide how to reduce AWS Secrets Manager and KMS costs.

Express is 25 times cheaper per invocation

The headline gap is stark: Standard workflows bill $25 per million state transitions, Express bills $1 per million requests. A workflow with even a handful of states run millions of times racks up tens of transitions per execution on Standard. If it is short-lived and does not need durable, exactly-once history, moving it to Express can cut its Step Functions line by more than 90 percent.

Should I use Step Functions Standard or Express workflows?

Use Express workflows for high-volume, short-duration orchestration and Standard workflows for long-running, durable processes that need exactly-once execution and a full audit history. The choice is the most important cost decision in Step Functions. Standard workflows are durable, can run up to a year, record every state in execution history, and guarantee exactly-once execution, which suits human-approval flows, long ETL pipelines, and anything that must be auditable. They bill $25 per million state transitions, so they are costly at high frequency. Express workflows run up to five minutes, bill $1 per million requests plus duration, and are built for high-throughput event processing where the occasional at-least-once retry is acceptable. The table below sets the two side by side.

FactorStandard workflowsExpress workflows
Price$25 per million state transitions$1 per million requests + duration
Max durationUp to 1 yearUp to 5 minutes
Execution semanticsExactly-once, full historyAt-least-once, high throughput
Best forLong, auditable, durable flowsHigh-volume short event processing

Verdict: high-volume short workflows belong on Express; reserve Standard for long-running, durable, audit-critical processes.

Are your workflows on the right billing model?

Our AWS cost audit reviews every state machine and queue, moves high-volume workflows to the cheaper model, batches messaging, and proves the saving against a clean baseline. On the performance model, you pay only from realized savings. No savings, no fee.

Book a cloud cost audit →
Go deeper · free field guide

The AWS Cost Optimization Field Guide includes the serverless checklist we run across Step Functions workflow types, SQS batching, and message volume.

Frequently asked questions

How much does AWS Step Functions cost?

AWS Step Functions Standard workflows cost $0.025 per 1,000 state transitions, which is $25 per million, with 4,000 transitions free per month. Express workflows cost $1 per million requests plus $0.06 per GB-hour of duration. For high-volume, short workflows, Express is dramatically cheaper than Standard.

How much does Amazon SQS cost?

Amazon SQS Standard queues cost $0.40 per million requests and FIFO queues cost $0.50 per million, with the first 1 million requests free each month. Each 64 KB chunk of a payload counts as one request, and batching up to 10 messages per call plus long polling cuts the request count.

Should I use Step Functions Standard or Express workflows?

Use Express workflows for high-volume, short-duration orchestration and Standard for long-running, durable workflows that need exactly-once execution and full audit history. Express is $1 per million requests while Standard is $25 per million state transitions, so high-throughput workloads are far cheaper on Express.

What is the biggest hidden Step Functions cost?

State transitions from retries, Wait states, and chatty workflows on Standard. Every retry and every step bills as a transition at $25 per million, so a workflow with many small states run at volume costs far more than the same logic on Express or with fewer states.

The short version

Step Functions Standard bills $25 per million state transitions and Express $1 per million, while SQS is $0.40 per million requests. Move high-volume short workflows to Express, cut transitions per execution, batch SQS messages with long polling, and use Standard queues where ordering is not required. When you want every workflow and queue reviewed and the saving proven down, that is part of what our AWS cost optimization service delivers.

Primary sources & further reading

Cloud pricing and service behavior change frequently. Verify the specifics in this guide against the providers’ own current documentation and the FinOps Foundation: AWS pricing ↗, AWS documentation ↗ and FinOps Foundation Framework ↗. This article also reflects Cloud Cost Room’s hands-on, vendor-neutral engagement experience.

Written by Morten Andersen

Co-founder of Cloud Cost Room and a FinOps Certified Practitioner, with 20 years in IT and cloud cost optimization across AWS, Azure, Google Cloud and OCI. More about Morten →

More from the AWS Cost Optimization cluster

See every guide in the AWS Cost Optimization cluster →

The Cloud Cost Brief

Cloud pricing moves. We tell you when it matters.

New commitment instruments, FOCUS changes, hyperscaler pricing shifts, and the plays that actually move a bill. No schedule, no filler.

Subscribe · Work email only