LMNT Amazon Seller backfill strategy
Date: March 19, 2026 Source: Granola Meeting ID: 77a5cc84-a5e4-48de-b58f-da20a197d15e URL: https://notes.granola.ai/t/77a5cc84-a5e4-48de-b58f-da20a197d15e
Participants:
- Uttam Kumaran (note creator) from Brainforge uttam@brainforge.ai
- Ghalib Suleiman from Polytomic ghalib@polytomic.com
Amazon Seller API Performance Issues
- Orders table sync running incrementally for 2026 data
- API extremely slow - took ~1 week for Jan 1-Mar 18 orders data
- Order items table particularly problematic
- Estimated 2-3 months just for 2026 data
- Would take months for full historical backfill
Required Tables & Constraints
- Priority tables identified:
- Orders (currently syncing, manageable timeframe)
- Order items (major bottleneck)
- Financial shipment event item
- Financial fee component (Amazon fees analysis)
- Order item promotion ID (discount codes)
- Need 6-month lookback minimum to start showing data
- Full historical backfill less critical than recent data access
Proposed Solution Strategy
- Contact Amazon Seller support for bulk data export
- Request historical dump via FTP or file transfer
- Precedent exists - other vendors provide this workaround
- Would bypass API rate limits for historical data
- Polytomic handles ongoing incremental updates post-backfill
- Client (LMNT) prefers detailed technical explanations in requests
Rate Limiting Analysis
- Ghalib to run tests on new table list
- Will provide specific rate limit numbers and runtime estimates
- Comprehensive report for Amazon support request
- Timeline: Early next week for full analysis
Next Steps
- Uttam: Draft detailed support request to Amazon with technical justification
- Ghalib: Deliver comprehensive table analysis with rate limits by early next week
- Parallel work continues on Walmart integration
- Follow up on Amazon bulk export feasibility