What are Marketing Cloud API rate limits, and how do you handle them in integration design?

Marketing Cloud enforces API rate limits per MID (business unit) to protect platform stability:
REST API: Generally 2,000 requests/minute. Some endpoints (e.g., contacts) have lower per-endpoint limits.
SOAP API: 2,000 requests/minute.
Transactional Messaging: Up to 2,500 emails/minute on dedicated IPs; shared IP limits are lower.
OAuth token requests: Throttled — cache tokens and reuse until near expiry (18 minutes).
Handling strategies:

1. Implement exponential back-off with retry on HTTP 429 responses.
2. Batch API calls where endpoints support it (e.g., TM API batch of 50).
3. Use the async data extension API for bulk DE operations rather than row-by-row REST calls.
4. Monitor rate limit headers (X-RateLimit-Remaining) in responses.
5. Distribute integrations across business units where architecture permits.