E-commerce intelligence diagnostic
Find ad waste before building the whole dashboard.
A diagnostic for Amazon Ads, ClickUp catalog data, Keepa/BSR signals, Claude summaries, Supabase metrics, and Slack Block Kit action reports.
First slice I would prove
| Workflow | Concrete output |
|---|---|
| Ad Waste Detector | Campaign/ad group/keyword rows with 10+ clicks, zero orders, ACoS above target, and estimated dollar impact. |
| Catalog Join | ClickUp ASIN/design status joined to ads rows with missing-catalog and stale-status exceptions separated. |
| Pricing Signal | Keepa/BSR + price + royalty math, with low-price/high-traction and high-price/dead-stock flags. |
| Slack Actions | Three ranked action blocks: stop waste, scale winner, inspect pricing. Each item needs a reason and dollar estimate. |
| Weekly Memory | Supabase tables for weekly baselines, rerun-safe imports, and week-over-week CEO brief context. |
Diagnostic deliverable
- Data contract for Amazon Ads, ClickUp, Keepa, royalty CSV, Supabase, Claude, and Slack.
- Risk list: OAuth refresh, pagination, API rate limits, missing ASINs, duplicate campaigns, and Claude hallucination guardrails.
- Phase-1 implementation map for the Ad Waste Detector only, before pricing and CEO brief complexity.
- Acceptance checks: sample rows in, Slack action blocks out, no duplicate alerts on rerun.
I do not need live credentials for the diagnostic. Redacted sample rows and target Slack examples are enough.
Request diagnostic