generated: '2026-08-12'
method: searched
source: https://pulsar.channel99.com/docs/#/
docs:
- https://pulsar.channel99.com/docs/#/
- https://support.channel99.com/hc/en-us/articles/49766041989787-Channel99-Reporting-API-Developer-Guide
derived_from: openapi/channel99-pulsar-openapi.json

note: >-
  Channel99 publishes its numeric limits inside the OpenAPI document itself - the
  components.responses.TooManyRequests description enumerates four distinct tiers and names the
  environment variable that disables the application-level tier. That is unusually specific for
  a B2B martech API. The prose developer guide is weaker, deferring to "the Swagger documentation
  or your Channel99 contract", so the spec is the authoritative source recorded here.

limits:
- scope: per-client_id
  applies_to: POST /auth/token
  limit: 20
  window: 1 minute
  enforced_at: application
  note: A separate WAF per-IP limit also applies to token issuance.
- scope: per-client
  applies_to: /events/* and /dimensions/*
  limit: 100
  window: 10 seconds
  burst: true
  enforced_at: application
  note: Described in the spec as a burst limit.
- scope: per-x-client-id
  applies_to: all authenticated routes
  limit: 60
  window: 1 second
  measured_over: 5 minutes
  enforced_at: waf
- scope: per-client
  applies_to: /ip/* (Channel99 IP API)
  limit: null
  window: null
  enforced_at: waf
  tier: elevated
  note: >-
    The spec states /ip/* routes get a WAF elevated tier and that the application burst limit
    does not apply. The numeric ceiling of the elevated tier is not published.

response:
  status_on_exhaustion: 429
  headers:
  - name: Retry-After
    description: Number of seconds to wait before retrying.
    type: integer
    example: 60
  body: '{"error":"...","code":"...","message":"Rate limit exceeded"}'
  standard_ratelimit_headers: false
  standard_ratelimit_headers_note: >-
    Channel99 does not return RateLimit-Limit / RateLimit-Remaining / RateLimit-Reset or the
    X-RateLimit-* family. Retry-After on a 429 is the only runtime signal, so a client cannot
    see how much budget remains before it is cut off - it can only react after the fact.

configuration:
- key: PULSAR_APP_RATE_LIMITS
  values: '0 | false | off'
  effect: Disables the application-level limits. WAF limits still apply.
  note: Published in the spec; presumably a deployment-side toggle rather than a customer control.

guidance:
  published: true
  source: https://support.channel99.com/hc/en-us/articles/49766041989787-Channel99-Reporting-API-Developer-Guide
  recommendations:
  - Use exponential backoff for 429 responses.
  - Avoid parallel requests over very large date ranges.
  - Cache lookup data such as channel, vendor, campaign and audience IDs.
  - Schedule large exports during off-peak hours where possible.
  - Retry failed pages rather than restarting the full export.

limit_count: 4