n8n Workflow: Google Sheets Catalog to Product Posts

Turn your approved Google Sheets catalog into ready product posts across 2–4 platforms

This n8n workflow converts catalog facts from Google Sheets into draft product posts for multiple platforms—validating SKUs, prices, store links, and content before you review exact text in an authenticated form. It also produces an optional scheduler CSV without auto-publishing.

What this workflow does

• Ingests catalog data: Reads up to 1,000 rows, rejects duplicate product IDs, invalid prices, unsupported currencies, unapproved content, and links outside your exact trusted store hosts. Selects up to three ready, previously unapproved product snapshots.
• Loads an authoritative pipeline record: Uses exactly one versioned pipeline entry from an n8n Data Table, then stores an exclusive request intent via compare-and-swap to prevent conflicting runs. It makes one OpenAI request with automatic HTTP retries disabled.
• Generates drafts with strict controls: Applies a strict JSON schema to select a neutral hook and one or two evidence IDs for each requested product/platform pair. The model is constrained to not write prices, discounts, availability claims, or URLs.
• Builds final drafts from source evidence: Drafts are assembled from selected exact source sentences, product title, catalog price and currency, catalog stock status, and the approved product URL. Validates platform length, missing pairs, unknown evidence, refusals, and incomplete responses.
• Saves outputs for traceability: Writes posts, evidence snapshots, and a content hash back to the Data Table, and mirrors individual drafts to Google Sheets using RAW values (the mirror is not an approval authority).
• Review companion: Includes an authenticated review companion that lets you review exact generated content. It also supports downloading an optional scheduler CSV—without publishing automatically.

Use cases

• Auto-drafting product announcements from an approved Google Sheets catalog for recurring marketing cycles.
• Consistent brand-safe messaging using evidence-based sentences and controlled OpenAI output.
• Pre-scheduling posts for multiple channels while keeping human approval in the loop.

Technical details

• Workflow/logic: if, set, code, Data Table, sticky note, compression
• Core integrations: Google Sheets (input + RAW mirroring) and OpenAI (single constrained request)
• Reliability controls: versioned pipeline record, compare-and-swap intent lock, and HTTP retry behavior disabled