specification: API Commons Rate Limits specificationVersion: '0.1' schema: https://raw.githubusercontent.com/api-evangelist/interface-research/main/schema/api-commons.yml#/$defs/RateLimits provider: Optimizely providerId: optimizely created: '2026-05-04' generated: '2026-08-13' method: searched modified: '2026-08-13' reconciled: true tags: - Rate Limiting - Experimentation - Customer Data description: >- Published API rate limits across the Optimizely estate. Optimizely documents limits in prose on two product docs pages (ODP and CMP) and nowhere else — and it returns NO rate-limit response headers anywhere: not one of the 28 harvested OpenAPI documents declares X-RateLimit-*, RateLimit-* or Retry-After, and only 2 of 1,606 documented error responses declare a 429 at all. An agent therefore cannot read remaining quota from an Optimizely response and must pace client-side against the numbers below. sources: - https://docs.developers.optimizely.com/optimizely-data-platform/reference/rate-limits - https://docs.developers.optimizely.com/content-marketing-platform/docs/rate-limits-1 - https://docs.developers.optimizely.com/platform-optimizely/docs/error-handling-and-rate-limits headers: {} headers_note: >- No rate-limit signalling headers are published or declared in any spec. Verified by grepping all 28 documents in openapi/_original/ on 2026-08-13. responseCodes: throttled: 429 limits: - name: ODP — Events API api: Optimizely Data Platform v3 scope: account metric: requests_per_second value: null unlimited: true note: Documented as "No limit". Bulk loads must still be batched (500 objects per request). source: https://docs.developers.optimizely.com/optimizely-data-platform/reference/rate-limits - name: ODP — Objects API api: Optimizely Data Platform v3 scope: account metric: requests_per_second value: null unlimited: true source: https://docs.developers.optimizely.com/optimizely-data-platform/reference/rate-limits - name: ODP — Lists API api: Optimizely Data Platform v3 scope: account metric: requests_per_second value: 10 source: https://docs.developers.optimizely.com/optimizely-data-platform/reference/rate-limits - name: ODP — Profiles API api: Optimizely Data Platform v3 scope: account metric: requests_per_second value: 10 note: >- Combined with 500-object batching this is the documented ceiling of 5,000 customer updates per second. source: https://docs.developers.optimizely.com/optimizely-data-platform/reference/rate-limits - name: ODP — Recommendations API api: Optimizely Data Platform v3 scope: account metric: requests_per_second value: 100 source: https://docs.developers.optimizely.com/optimizely-data-platform/reference/rate-limits - name: ODP — On-Site GraphQL api: Optimizely Data Platform v3 scope: account metric: requests_per_second value: 500 source: https://docs.developers.optimizely.com/optimizely-data-platform/reference/rate-limits - name: CMP Open REST API api: Content Marketing Platform v3 scope: organization metric: requests_per_second value: 10 source: https://docs.developers.optimizely.com/content-marketing-platform/docs/rate-limits-1 - name: CMP monthly request quota api: Content Marketing Platform v3 scope: organization metric: requests_per_month value: 200000 note: Optimizely Support can raise or reset the monthly quota. source: https://docs.developers.optimizely.com/content-marketing-platform/docs/rate-limits-1 undocumented: - api: Experimentation REST API v2 (api.optimizely.com/v2) note: >- No rate-limit page is published for the Web/Feature Experimentation REST API on the developer portal as of 2026-08-13. A previous API Evangelist sweep recorded "100 requests/minute per project" sourced from optimizely.com/plans/; that number could NOT be re-verified against any current provider page in this pass and is therefore recorded here as unverified rather than as a published limit. - api: Feature Experimentation Flags v1 (api.optimizely.com/flags/v1) - api: Optimizely Campaign REST API - api: Configured Commerce Admin / Storefront - api: Optimizely Graph note: >- The Graph docs page "Error handling and rate limits" describes that rate limits exist and restrict query volume or complexity, but publishes no numbers. limit_count: 8