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: Gladly
providerId: gladly
created: '2026-05-24'
modified: '2026-05-24'
reconciled: true
tags:
  - Rate Limiting
  - Quotas
  - Customer Service
description: 'Reconciled rate limits for the Gladly REST API. Default 10 requests per second per organization across all HTTP methods, with separate Reporting API tier and per-second response headers.'
sources:
  - https://developer.gladly.com/rest/
  - https://help.gladly.com/developer-tutorials/docs/default-api-rate-limits
headers:
  limit: Ratelimit-Limit-Second
  remaining: Ratelimit-Remaining-Second
  retryAfter: Retry-After
responseCodes:
  throttled: 429
  quotaExceeded: 429
algorithm: fixed-window-per-second
limits:
  - tier: Default REST API
    scope: organization
    rpm: 600
    rps: 10
    burst: 10
    methods:
      - GET
      - POST
      - PUT
      - PATCH
      - DELETE
    notes: 10 requests per second per organization across all HTTP methods. Concurrency not enforced separately. Exceeding the limit returns HTTP 429.
  - tier: Reporting API
    scope: organization
    rps: 'documented separately'
    notes: The Reporting API has its own dedicated tier described in the developer portal; aggregate report jobs rather than polling at the default 10 req/sec.
recommendations:
  - Honor Ratelimit-Remaining-Second to back off proactively.
  - For bulk loads, use the Export jobs API instead of looping over per-record endpoints.
  - For event ingestion, prefer Webhooks over polling.
  - On 429, retry with exponential backoff starting at 1s.