Skip to main content

​
Overview

Rather than fetching all merchants on every sync, use updated_since to pull only records that changed since your last sync. This reduces API calls and keeps you well within rate limits.

​
Full sync (first run)

On your first sync, page through all merchants: 
import requests

BASE = "https://modulate.aurorapayments.net/api/v2/partner"
HEADERS = {"X-Partner-API-Key": "pk_live_YOUR_KEY_HERE"}

def full_sync():
    merchants = []
    page = 1

    while True:
        resp = requests.get(
            f"{BASE}/merchants",
            headers=HEADERS,
            params={"page": page, "page_size": 200}
        ).json()

        merchants.extend(resp["data"])
        last_sync = resp["meta"]["as_of"]

        if page >= resp["meta"]["pagination"]["total_pages"]:
            break
        page += 1

    return merchants, last_sync

​
Incremental sync (subsequent runs)

Pass updated_since with the as_of timestamp from your last sync: 
def incremental_sync(last_sync_timestamp):
    updated = []
    page = 1

    while True:
        resp = requests.get(
            f"{BASE}/merchants",
            headers=HEADERS,
            params={
                "page": page,
                "page_size": 200,
                "updated_since": last_sync_timestamp
            }
        ).json()

        updated.extend(resp["data"])
        new_sync = resp["meta"]["as_of"]

        if page >= resp["meta"]["pagination"]["total_pages"]:
            break
        page += 1

    return updated, new_sync
Portfolio sizeSync frequencyEstimated calls/day
< 1,000 merchantsEvery 4 hours~6
1,000–5,000Every 2 hours~24
5,000+Every hour~48
Incremental syncs typically return a small fraction of your portfolio, so even hourly syncs stay well within the 120 req/min rate limit.

​
Tips

  • Store as_of from each sync response as your next updated_since value
  • Upsert, don’t replace — Incremental results contain only changed records
  • Filter by status — Use ?status=active if you only care about active merchants
  • Handle pagination — Even incremental results can span multiple pages after bulk updates