Plan traffic and handle throttling
Numeric quota: not yet confirmed for publication.
A verified per-second/per-minute limit, burst allowance and quota scope (IP, key or company) are not currently available in this public contract. Do not assume unlimited traffic or use a limit reported for another integration. Ask support to confirm capacity for your environment and expected peak traffic before a bulk import.
A 429 Too Many Requests response means requests are being throttled. Reduce concurrency and spread requests over time, including status polling.
- If the response includes
Retry-After, wait for the indicated delay or date before retrying. - If it is absent, use exponential backoff with jitter. For read requests, a suggested starting schedule is approximately 1, 2, 4, 8, 16 and 30 seconds; cap the attempts and surface an actionable error.
- For a creation request, follow the safe re-submission flow before retrying. Do not assume every error means no assignment was created.
Retry-After and rate-limit headers are not guaranteed. Errors from infrastructure can also have a different body from application errors. The backoff schedule above is client guidance, not a published throughput allowance.