# Rate Limiting

> Rate limiting is a control that caps how many requests a client can make to an API within a period of time, such as per second, per minute or per day. When a client exceeds the limit, the API rejects further requests, usually with an HTTP 429 Too Many Requests error, until the window resets. AI model providers also limit tokens processed per minute.

## Why it matters for your business

Rate limits are a common reason automations work in testing and fail in production. A backfill of 20,000 records, a burst of orders on a sale day, or an AI agent making many tool calls can all hit a limit and leave work half done. Well-built integrations read the limit headers, slow down and retry with backoff, queue work during spikes and resume where they stopped. Limits often differ by plan tier, so check them before choosing a plan or promising a processing speed.

## Example

A company migrates five years of customer records from an old CRM to a new one. The first import script sends requests as fast as possible and is blocked by the new CRM's rate limit within minutes. The rewritten script processes records in batches, pauses when it receives a 429 response, retries with increasing delays and logs progress so it can resume after any interruption.

Related terms: https://www.lumeor.studio/glossary/api-application-programming-interface/, https://www.lumeor.studio/glossary/idempotency/, https://www.lumeor.studio/glossary/webhook/, https://www.lumeor.studio/glossary/observability/

---

Source: https://www.lumeor.studio/glossary/rate-limiting/