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: Leanplum
providerId: leanplum
created: '2026-07-03'
modified: '2026-08-13'
generated: '2026-08-13'
method: searched
source: https://docs.leanplum.com/reference/responses
reconciled: false
tags:
- Mobile Marketing
- Customer Engagement
- Messaging
- Rate Limiting
- Quotas
description: >-
  Leanplum publishes no global per-minute request cap and no rate-limit response
  headers, but it does publish four hard, numeric constraints: a 1 request/second
  per-device cap on development-mode traffic, a concurrency limit of one in-flight
  request per userId ("strict device locking"), the multi batching limits of 50
  users / 500 actions per call, and a data-export quota of 24 exports per day. The
  only runtime throttling signal is the status code (429 for the rate and
  concurrency limits, 403 for an oversized multi) plus a message string inside
  response[].error.message — a client cannot read remaining quota anywhere.
notes: >-
  Upgraded 2026-08-13 from the published Responses reference, which documents the
  devMode 1 rps limit, strict device locking, the 429 status, and the recommended
  backoff. No numeric per-account request-rate limit is published; throughput beyond
  these caps is governed by contract terms. Design server-side integrations to batch
  per user first through multi, never issue concurrent calls for the same userId,
  and retry 429/5xx with exponential backoff.
sources:
- https://docs.leanplum.com/reference/responses
- https://docs.leanplum.com/reference/batching-requests
- https://docs.leanplum.com/reference/reduce-billable-requests
- https://docs.leanplum.com/reference/making-requests
- https://docs.leanplum.com/reference/get_api-action-exportdata
- https://docs.leanplum.com/reference/post_api-action-addpostback
responseCodes:
  throttled: 429
  batch_exceeded: 403
headers:
  published: false
  request: []
  response: []
  retry_after: false
  notes: >-
    Leanplum publishes no X-RateLimit-*, no RateLimit-*, and no Retry-After header.
    The client learns it was throttled only from the status code and the human-readable
    string in response[].error.message ("Per device rate limit exceeded.",
    "Too Many Requests.", "Request failed due to concurrent requests to the same
    profile ID").
limits:
- name: Development Mode - Per Device
  scope: device
  metric: requests
  limit: 1
  window: second
  responseCode: 429
  errorMessages:
  - Per device rate limit exceeded.
  - Too Many Requests.
  notes: >-
    Applies to any request carrying devMode=true. Published at
    https://docs.leanplum.com/reference/responses.
- name: Strict Device Locking - Per Profile Concurrency
  scope: user
  metric: concurrent requests
  limit: 1
  window: concurrent
  responseCode: 429
  errorMessages:
  - Request failed due to concurrent requests to the same profile ID
  notes: >-
    No two GET or POST requests may execute at the same time against the same userId;
    Leanplum serializes them to keep data consistent and fails the loser. Fold per-user
    work into a single multi call.
- name: Multi Batch - Users
  scope: request
  metric: users
  limit: 50
  responseCode: 403
  notes: A single multi call may reference at most 50 users; exceeding this returns 403 and the call is ignored.
- name: Multi Batch - Actions
  scope: request
  metric: actions
  limit: 500
  responseCode: 403
  notes: A single multi call may contain at most 500 actions; exceeding this returns 403 and the call is ignored.
- name: Data Export - Daily Jobs
  scope: account
  metric: exports
  limit: 24
  window: day
  notes: >-
    exportData may be run at most 24 times per day. Exports with invalid arguments do
    not count toward the limit.
- name: Data Export - Lookback Window
  scope: account
  metric: days
  limit: 60
  notes: You cannot export data that became available more than 60 days ago.
- name: Data Export - Result Retention
  scope: request
  metric: hours
  limit: 24
  notes: Export files are automatically deleted 24 hours after the export occurs.
- name: Postbacks Per App
  scope: account
  metric: postback rules
  limit: 3
  notes: The maximum number of registered postbacks is three per app.
- name: GET Request URL Length
  scope: request
  metric: characters
  limit: 2083
  notes: GET requests are bounded by the ~2,083 character URL limit; use POST for larger payloads.
- name: Account Throughput
  scope: account
  metric: requests
  limit: not published
  notes: No fixed numeric per-account request-rate limit is documented; throughput is governed by contract terms.
policies:
- name: Retry with exponential backoff
  description: >-
    On 429 or 5xx, Leanplum's published guidance is to retry after 2 ± random(0,1)
    seconds, then 4 ± random(0,2), then 8 ± random(0,4), up to a bounded retry count.
  source: https://docs.leanplum.com/reference/responses
- name: Billable User Lookups
  description: >-
    Each unique user lookup in a request is one billable API call. Batching per user
    first and then across users controls both throughput and cost; batching by action
    instead doubles the billable count for the same work.
  source: https://docs.leanplum.com/reference/reduce-billable-requests
- name: Concurrent Request Queuing
  description: >-
    While a multi batch is processing, concurrent requests for the same users (from
    SDK or API) are queued and executed after the batch completes.
- name: Postback Delivery Backoff
  description: >-
    Outbound postbacks time out after 30 seconds; on a 5xx from the receiver Leanplum
    retries up to 9 more times with exponential backoff starting at 1 hour, up to 10
    hours, after which the data is lost.
  source: https://docs.leanplum.com/reference/post_api-action-addpostback
maintainers:
- FN: Kin Lane
  email: kin@apievangelist.com