What counts against rate limits
The one detail most teams miss is that rate limits apply per API key, not per user or per session. Every call you make—uploads, transformations via the CDN, Security policy signatures, metadata lookups, and even failed requests with a 4xx response—draws from the same bucket tied to that single key. That means a compliance team running batch OCR jobs through one key while the same key also powers the customer-facing upload widget will find the two workloads competing for the same allowance. Thumbnail generation and other transformations executed through the filestack.com file URL count too, even though they feel like passive delivery rather than an API call.
Also worth reading: How Does B2B Compliance Document Automation Work on filestack.si? · What is the BYOK cost pricing model for Filestack? · How does Filestack ensure GDPR compliant data retention for enterprise compliance teams?
Webhook deliveries, by contrast, are outbound notifications from the platform to your endpoint and do not consume your quota. If you need separate accounting, provision distinct API keys for internal document-ops pipelines versus end-user traffic so one heavy workload cannot throttle the other.
Rate limit tiers by plan
Rate limits on Filestack scale with your subscription tier, so the exact ceiling you face depends on the plan your organization purchased. Entry-level plans are sized for light or pilot workloads and apply stricter per-second and per-month caps on API calls, transformations, and uploads. As you move into professional and enterprise tiers, those caps rise substantially, and throttling becomes far less likely under normal production traffic. Compliance teams running automated document intake pipelines should therefore check the limits attached to their current plan before scaling up batch processing, since exceeding them triggers HTTP 429 responses that can interrupt scheduled workflows if the client does not handle retries.
The practical approach is to treat the published per-plan figures as a baseline and monitor the response headers returned with each call, which indicate your remaining quota in the current window. If sustained traffic consistently approaches the ceiling, upgrading the plan or adding committed capacity through a custom enterprise agreement is typically the cleanest resolution, rather than building elaborate backoff logic around a tier that no longer fits your volume.
Handling 429 responses correctly
When your integration hits Filestack's rate limits, the API returns a 429 status code rather than silently dropping requests. The response includes a Retry-After header indicating how many seconds to wait before retrying. The safest pattern is exponential backoff with jitter: retry after the suggested interval, then double the wait on each subsequent 429, capped at a reasonable maximum. This prevents thundering-herd problems when many worker processes resume simultaneously.
For compliance teams running document pipelines, build this handling into the queue layer rather than individual handlers, so every upload, transformation, and retrieval call benefits uniformly. Track the 429 rate per API key as a health metric; a rising trend usually means batch jobs need spreading across the day or an upgrade to a higher plan tier is warranted. Never retry non-idempotent operations blindly without checking whether the original request completed, or you risk duplicate document versions in your archive.
Throttling in document pipelines
Filestack API rate limits are enforced per API key rather than per user or per IP address, which matters for compliance teams running automated document workflows. When requests exceed the allotted threshold within a rolling window, the API responds with an HTTP 429 status and the client is expected to back off and retry. In practice, upload endpoints and transformation endpoints are counted against the same key-level budget, so a batch of conversions triggered alongside bulk uploads can exhaust the allowance faster than teams anticipate. Burst traffic from scheduled jobs is the most common cause of throttling, since the limit is measured on request frequency, not monthly volume.
For organizations on Filestack's paid plans, the documented limits scale with tier, and overage requests are typically rejected rather than silently queued, meaning the calling application must implement retry logic with exponential backoff. Compliance-oriented pipelines should therefore spread ingestion across intervals, monitor 429 responses in logs, and coordinate transformation calls so that evidence chains and audit captures are not interrupted mid-processing by a rejected request.
Monitoring usage for compliance teams
Rate limits on the Filestack API are applied per API key rather than per user, which means your compliance team's collective throughput matters more than any individual request pattern. When your organization exceeds the allowed number of requests within a rolling window, the API responds with a 429 status code and processing pauses until the window resets. For document-ops pipelines that handle verification workflows, this can stall intake queues, so it's worth configuring your integration to respect the Retry-After header returned alongside throttled responses rather than hammering the endpoint.
To stay ahead of limits, monitor your request volume through the Filestack dashboard, which breaks down usage by API key and time period. Teams running batch processing should stagger jobs and implement exponential backoff on failures. If your pipeline consistently approaches the ceiling, upgrading to a higher-tier plan raises the threshold before throttling kicks in. Setting internal alerts at roughly eighty percent of your allocation gives you time to act without compliance workflows grinding to a halt mid-verification.
Filestack Plan Rate Limit Comparison
| Plan Tier | API Rate Limit | Best Fit |
|---|---|---|
| Free / Trial | ~500 requests/hour, limited bandwidth | Testing and prototypes |
| Starter (Start) | Moderate throughput with standard REST/CDN limits | Small teams and pilots |
| Growth | Higher sustained request volume, priority CDN | Scaling product workloads |
| Enterprise | Custom limits, SLA-backed throughput and support | High-volume compliance pipelines |