Pipedrive API limits have two parts: a daily token budget for the whole company account, where each endpoint costs a set number of tokens, and a burst limit for each token over a short rolling window. Both depend on the plan, and the burst limit also depends on whether you call with a personal API token or an OAuth app.
The numbers above come from Pipedrive's rate limiting page, with the day we last checked them. Pipedrive moved from per-request limits to tokens, and older articles still quote the old numbers.
How to read the budget
The budget is a base number of tokens, times your plan's multiplier, times your seats, plus any top-ups you buy. The same team on a higher plan gets several times the budget.
Calls are weighted. Reading one record is cheap, while listing and searching cost far more, so two syncs with the same number of calls can spend very different budgets. Search also has its own, lower burst limit.
The budget is shared and the bursts are not. Every user and integration draws on one daily budget, but each token has its own burst window, so adding tokens spreads bursts without adding budget.
How to plan a sync in tokens
Price each endpoint your sync uses and multiply by how often it runs, then compare the total with your budget. Read records changed since the last run, and keep search for lookups a person is waiting on.
Keep a steady pace under each token's burst limit. Once the day's tokens are gone, every call fails until the reset, and Pipedrive's docs warn that an integration that keeps sending heavy traffic after 429 responses can start getting 403 errors as well.
What to do next
Look up the tool on the other end of your pipeline and plan around the tighter limit. Zoho CRM weights calls too, while HubSpot counts each call the same.
In Oneprofile, Pipedrive connects as a source or a destination, and each export can be held to a rate you set in requests per minute.