---
updatedAt: 2026-04-09T10:34:03.000Z
agentTools:
  projectIndex: https://developer.productboard.com/llms.txt
---

# Rate limits

To ensure fair usage and platform stability, the Productboard API enforces rate limits on incoming requests.

## Limit

The default rate limit is **50 requests per second**, per access token

If you exceed this limit, the API will respond with:

```http
HTTP 429 Too Many Requests
```

This means you're sending requests too quickly and need to slow down before continuing.

***

## Rate Limit Headers

The API includes the following headers to help you track usage:

| Header                  | Description                                                     |
| ----------------------- | --------------------------------------------------------------- |
| `X-RateLimit-Limit`     | Maximum number of requests allowed in the current time window   |
| `X-RateLimit-Remaining` | Number of requests remaining before hitting the limit           |
| `Retry-After`           | (Only when throttled) Number of seconds to wait before retrying |

Use these headers to implement smart retry logic and avoid being throttled.

***

## Handling 429 Responses

When you receive a `429 Too Many Requests` response:

* Check for the `Retry-After` header, which tells you how many seconds to wait
* Back off and retry the request after the delay
* Avoid retrying immediately or in a tight loop

Example:

```http
HTTP/1.1 429 Too Many Requests
Retry-After: 2
```

In this case, wait 2 seconds before retrying.

***

## Best Practices

* **Avoid bursts**: Spread requests evenly across time rather than sending them all at once
* **Cache responses**: Especially for read-heavy use cases, cache API results locally to avoid unnecessary requests
* **Use rate limit headers**: Proactively monitor usage and implement logic to pause or back off when nearing the limit
* **Retry responsibly**: Use exponential backoff or delay-based retries when encountering 429s