Core concepts
Rate limits & concurrency
Two separate limits: concurrency (how many calls run at once, per account) and request rate (API calls per second, per key).
Concurrency
Concurrency is how many item calls (products, offers, offer checks) run at the same time across your account. It grows with the credits you have ever bought, and never goes back down.
| Lifetime credits purchased | Concurrent calls |
|---|---|
| Any | 5 |
| 25,000+ | 20 |
| 500,000+ | 50 |
| 2,000,000+ | 100 |
| 10,000,000+ | 250 |
- A call with no free slot fails at once:
429 CONCURRENCY_LIMITwithRetry-After: 1. Nothing is charged. - Watchlist runs are scheduled for you and do not take your slots.
- Each item call uses one slot however many ids it carries, so fewer, larger calls go further.
Check your limit
curl "https://api.scrapelyserp.com/v1/account" \
-H "Authorization: Bearer $SCRAPELY_API_KEY"concurrency holds your limit, the calls in_flight now and the next_tier.
Request rate
Separately, each API key may make 50 calls per second, counting every endpoint, including free ones such as listing watchlist runs. Going over is a 429 RATE_LIMITED.
| Header | Meaning |
|---|---|
X-RateLimit-Limit | Calls allowed per window. |
X-RateLimit-Remaining | Calls left in the current window. |
X-RateLimit-Reset | Unix time (seconds) when a call frees up. |
Retry-After | On a 429: seconds to wait. |
Use webhooks, not polling
Polling watchlist items in a tight loop is the usual way to hit the rate limit. Register a webhook endpoint and let us tell you when something changes.