name: Payflex API Rate Limits
description: >
  Payflex does not publish explicit rate limit figures in its public
  documentation. The API is designed for merchant-initiated e-commerce
  transactions (order creation, status checks, refunds) rather than
  high-frequency bulk calls. Standard REST best practices apply: implement
  exponential back-off on HTTP 429 or 503 responses, and avoid polling the
  order status endpoint more frequently than necessary — Payflex delivers
  order status updates via statusCallbackUrl webhook instead.
rate_limits:
  - scope: Authentication
    endpoint: "POST https://auth.payflex.co.za/auth/merchant"
    notes: >
      Access tokens are short-lived. Cache the token and reuse it until
      expiry rather than requesting a new token per API call. Excessive
      token requests may be throttled.
  - scope: Order Operations
    endpoint: "POST https://api.payflex.co.za/order/productSelect"
    notes: >
      No published limit. Intended for real-time checkout flows; one call
      per consumer checkout session is the expected usage pattern.
  - scope: Order Status Retrieval
    endpoint: "GET https://api.payflex.co.za/order/{orderId}"
    notes: >
      Prefer the statusCallbackUrl webhook for order status updates.
      Polling should be used only for reconciliation; poll infrequently
      (e.g., once per minute or less).
  - scope: Refunds
    endpoint: "POST https://api.payflex.co.za/order/{orderId}/refund"
    notes: >
      Idempotent when a unique requestId is supplied. Avoid duplicate
      refund submissions for the same orderId and amount.
  - scope: Configuration
    endpoint: "GET https://api.payflex.co.za/configuration"
    notes: >
      Returns merchant-specific payment limits. Cache this response;
      it rarely changes and does not need to be fetched on every request.
recommendations:
  - Implement OAuth2 token caching and refresh logic
  - Use the statusCallbackUrl webhook for order lifecycle events
  - Apply exponential back-off on 4xx/5xx responses
  - Contact Payflex support for volume-specific guidance on limits