Rate limits and pagination
How many requests you can make, the headers that tell you where you stand, and how to page through large collections.
Rate limits
Two limits apply, both in fixed one-minute windows. The workspace limit covers every API request of the workspace together (REST API with an API token, and the Zendesk-compatible API) and depends on your plan:
| Plan | Requests per minute, whole workspace |
|---|---|
| Free | No API access |
| Starter | 200 |
| Growth | 400 |
| Pro | 700 |
| Enterprise | 2,500 |
In addition, a single API token can make at most 700 requests per minute.
| Header | Meaning |
|---|---|
X-Rate-Limit, X-Rate-Limit-Remaining | The workspace limit and what is left of it in this minute |
X-RateLimit-Limit, X-RateLimit-Remaining | The same for the token |
X-RateLimit-Reset | Seconds until the token window resets |
Beyond the limit you get 429 Too Many Requests with a Retry-After header. Wait that many seconds, then retry. Back off exponentially if you keep getting 429.
Pagination
Collections are page-based: ?page=2&itemsPerPage=50, up to 100 items per page.
For exports and syncs, GET /api/tickets also supports cursor pagination, which stays consistent while tickets are created:
- Send
?cursor=&limit=100(an empty cursor) for the first page. - Follow
links.nextuntilmeta.has_moreis false.
Results are ordered by id, which is creation order. The Zendesk-compatible API supports the Zendesk offset, cursor and incremental export styles.