ERPNext Meta & TikTok Catalog Connector

ECOSIRE builds a proper Frappe app (installed on your bench, frappe/erpnext v15 or v16) that adds a Social Catalog Channel DocType per destination (a Meta Commerce catalog, a TikTok catalog, or both), plus a Social Catalog Item Map that resolves each ERPNext Item/variant to a channel product with its retailer_id, GTIN/MPN, brand, Google product category, and image set. A full feed is generated on demand and on a scheduler_events cron (typically hourly for a full reconcile), while incremental deltas fire from hooks.py doc_events — an on_update on Item, Item Price, and stock ledger changes queues a targeted push through frappe.enqueue so a price edit or a low-stock threshold reaches Commerce Manager and TikTok within minutes, not overnight. We push via the Marketing/Graph API (Catalog Batch endpoints) and the TikTok Business catalog API, store credentials in a masked Password field, and log every request/response, retry, and rejection in a Social Catalog Sync Log DocType so you can see exactly which SKU Meta refused and why.

Under the hood the connector handles the parts that break naive exports: variant_of templates map to variant groups with item_group_id, set/collection products are assembled from BOM or a configured grouping field, availability and inventory are derived from Bin actual/projected qty against a per-channel buffer, and multi-currency Item Price lists resolve to the right price/sale_price per catalog. A Promotion mapping emits sale price and effective-date windows so your ad creative reflects live campaigns. Everything is permission-scoped with a dedicated Role Profile, exposes whitelisted methods (@frappe.whitelist()) for a manual "resync now" client-script button on the Item form, and ships with a Frappe REST-friendly surface so your own automations can trigger a channel push. Field mappings, category taxonomies, and buffer rules are configuration, not hard-code — so your merchandisers adjust them without a developer.

Because this is build-to-order, we start from your real ERPNext instance rather than a generic template. A short scoping call confirms your item model (variants, sets, kits), which channels and catalogs you're feeding, your GTIN/brand data quality, and your currency/region matrix. We then build against a staging copy of your data, run UAT with your team pushing to a sandbox catalog, and only cut over to production catalogs once feeds validate clean. Typical delivery is 2-4 weeks from confirmed scope, and you receive the full source, a git repo, technical and user documentation, training, and a post-go-live support window — you own it, and it runs on your bench.