generated: '2026-08-26'
method: searched
source: https://moz.com/api/docs/guides/error-handling
docs: https://moz.com/api/docs/guides/error-handling
pricing_docs: https://moz.com/products/api/pricing
summary: >-
  Moz meters the API two ways: a monthly row-based quota tied to the subscription tier, and a
  penalty rate limit that only engages after repeated 4xx responses. The free tier additionally
  carries a hard request-rate cap of one request every ten seconds.
limit_count: 3
rate_limits:
- id: error-penalty-rate-limit
  scope: per-key
  window: 5 minutes
  limit: 50
  trigger: >-
    Engages once more than 50 requests in a 5-minute period return a 4xx error code. Once engaged,
    the caller is limited to 50 total requests per 5-minute period.
  reset: The limit is lifted after 30 seconds without receiving a 4xx response.
  source: https://moz.com/api/docs/guides/error-handling
- id: free-tier-request-rate
  scope: per-account
  window: 10 seconds
  limit: 1
  applies_to: Free tier
  note: '"Free access allows for one request every ten seconds, up to 50 rows per month."'
  source: https://moz.com/products/api/pricing
- id: monthly-row-quota
  scope: per-account
  window: 1 month
  limit: varies by tier (50 rows free, up to 4,000,000 rows on Advanced)
  unit: rows
  note: >-
    Quota is measured in rows - typically one row per object in a multi-object response, or one row
    per field for a flat single-object response. Consumption is reported in the response body of
    every call, and remaining quota is readable with the quota.lookup method
    (path api.limits.data.rows). Beta methods draw on a separate beta quota that does not reduce the
    main plan or incur overages.
  source: https://moz.com/api/docs/guides/getting-started
response_headers:
  observed: []
  note: >-
    Probed anonymously 2026-08-26 against https://api.moz.com/jsonrpc. No X-RateLimit-*,
    RateLimit-* or Retry-After headers were present on 200/400/401 responses. Moz surfaces quota
    consumption in the JSON response body instead of in headers, which means an agent cannot detect
    remaining budget or throttle state from response metadata alone.
  headers_present_unrelated: [x-service-name, x-service-version, x-pylon-action]
exhaustion:
  status_code: 4xx (the error-handling guide does not name a specific status for quota exhaustion)
  quota_lookup_method: quota.lookup
  overages: >-
    Some tiers include pay-as-you-go overages ($5 per 1,000 rows on Starter Large; $25/$20/$10 per
    10,000 rows across the Growth tiers; $5 per 10,000 rows on Advanced). Starter Medium has no
    overage option.