In plain language
It protects shared capacity and may require an automation to slow down or wait before trying again.
Example
An API permits 100 requests per minute and returns HTTP 429 when a batch exceeds that allowance.
Related terms
Common misconception
Staying below an advertised average does not guarantee every request succeeds because burst limits and other quotas may also apply.
Implementation guidance
Read response headers, queue bursts, use exponential backoff with jitter, cap retries, and request only the data you need.