Limits

​

Request rates

import time, requests

def get_with_retry(url, headers, attempts=5):
    delay = 1.0
    for attempt in range(attempts):
        resp = requests.get(url, headers=headers)
        if resp.status_code != 429:
            resp.raise_for_status()
            return resp.json()
        wait = float(resp.headers.get("Retry-After", delay))
        time.sleep(wait)
        delay = min(delay * 2, 30)          # fallback when the header is absent
    raise RuntimeError("rate limit did not clear in 5 attempts")

• keep your rate at 8–9 requests per second to leave headroom for retries;
• fetch pages at the maximum size (limit=1000) rather than in small chunks;
• on a 429, stop sending parallel requests — wait out the window.

​

Lifetimes

​

Webhook delivery

• Each event gets up to 8 delivery attempts with exponentially growing intervals, roughly a day in total.
• Success is any 2xx response within a reasonable time; everything else (timeout, 3xx, 4xx, 5xx) counts as a failure and is retried.
• After 20 consecutive permanently failed deliveries the subscription is automatically disabled. Re-enable it in the app (Settings → Integrations → “For developers” → Webhooks); there is currently no re-enable endpoint in the API.

​

Sizes

• An Ingest batch has no hard record cap, but keep it within a few thousand records: processing is asynchronous, and the request body should stay a reasonable size (single-digit megabytes).
• List responses are bounded by pagination: limit defaults to 100, maximum 1000.