Rate Limits by Plan
Each organisation has a token bucket that is refilled every minute. The bucket size depends on your plan:
In addition, a global limit of 10,000 requests per minute applies per IP address. There is no concurrency limit.
Rate-limited requests are not billed. The
/me endpoint is not rate limited.
Reasons and How to Fix
Too Many Requests
You’re sending requests faster than your plan’s per-minute limit allows. To fix:- Implement exponential backoff in your code
- Spread requests evenly over the minute rather than sending them in a burst
- Upgrade to a higher plan for a larger per-minute limit
No Retry-After Header
The API does not send aRetry-After header. Use exponential backoff with jitter and retry; the bucket refills within a minute.
Best Practices:
Batching in Workflow Tools
In n8n, Zapier or Make, keep each batch at or below your plan’s per-minute limit and wait 60 seconds between batches. For example, on the Free plan send 4 requests, then wait 60 s; on Pro send up to 12, then wait 60 s.Need Help?
Contact us at support@poof.bg with yourrequest_id if you need higher limits.