What a Recovery Baseline actually returns.
This example uses invented catalog, sales, inventory, and channel-issue exports. It demonstrates the operating method without claiming a customer relationship, lost revenue, or a guaranteed recovery. The Baseline stops at a controlled decision package; later Sprint artifacts are labeled separately and are not included in the Baseline.
Recovery Baseline
Executive decision
Proceed with a controlled sprint on five products; quarantine two identity conflicts first. The catalog contains broad content gaps, but the queue concentrates work where verified severity overlaps with supplied demand and inventory. Do not rewrite the full catalog.
Financial boundary: historical sales attached to affected products are observed exposure, not sales proven lost. Modeled upside is withheld until the comparison group, measurement window, availability, promotion, and margin assumptions are agreed.
Recovery Baseline
Input and identity coverage
| Source | Records | Join result |
|---|---|---|
| Shopify product export | 11 products / 12 variants | Canonical handles and variant SKUs established |
| 90-day sales export | 11 performance rows | 10 exact matches; one duplicate-SKU row excluded from financial totals |
| Channel-issue export | 3 issues | Two explicit blocks and one limited-status warning attached to products |
| Asset evidence | 11 image references | One missing hero asset and one duplicate-SKU conflict isolated |
Recovery Baseline
Verified exposure map
| Product | Observed evidence | Commercial signal | Confidence |
|---|---|---|---|
| GUEST-PNK-01 | Non-descriptive title + supplied Google limitation | $9,100 historical sales / 5,200 sessions | High · exact SKU |
| CAN-AMBER-8 | Missing product type + supplied Faire ineligibility | $7,800 historical sales / 210 units in stock | High · exact SKU |
| TOTE-NAT-01 | All-caps, thin context, incomplete search fields | $5,400 historical sales / 2,800 sessions | High · exact SKU |
| BOARD-OLV-01 | Missing product media + supplied Shopify error | $4,600 historical sales / 18 units in stock | High · exact SKU |
| TOW-BLU-01 | SKU collision across two product records | $2,100 supplied performance row excluded from automated exposure | Exception · human decision |
Recovery Baseline
Commercial recovery queue
| Priority | Why now | Controlled next action |
|---|---|---|
| 1 | Explicit channel restriction + highest exposed sales | Confirm product identity and approved product type; prepare grounded title diff |
| 2 | Faire ineligibility + high sales + deep inventory | Resolve taxonomy from approved source; validate destination rule before export |
| 3 | Missing production media on a selling product | Match or request the approved hero asset; block import until verified |
| 4 | Demand signal overlaps with incomplete buyer context | Draft from approved facts and show every source behind the change |
| HOLD | Duplicate identity creates overwrite risk | Product owner chooses the canonical SKU relationship; automation does not guess |
Recovery Baseline
Approved-fact change brief
| Field | Before | Proposed | Evidence state |
|---|---|---|---|
| Title | GLOW | Draft withheld | Product type and material still need authoritative confirmation |
| Product type | Blank | Candle | Requires customer approval; channel issue indicates missing type |
| Hero image | Blank | BOARD-OLV-01_HERO.jpg | Only after checksum and product-owner match approval |
| AI provenance | Not applicable | Recorded per generated field | Facts used, source, rule profile, model/version, and approval retained |
Verified Recovery Sprint · after a successful Baseline
Safe release pack
This is a Sprint deliverable, not a Baseline deliverable. It includes a minimal field-level diff, excluded/ambiguous records, rule validation, customer approval status, an import-ready CSV for the chosen transport, a pre-change backup, rollback instructions, and a post-import verification checklist. Blank or dependent fields are not included casually.
Verified Recovery Sprint · after a successful Baseline
30-day recovery ledger
This is a Sprint measurement deliverable. Freeze product eligibility, impressions, sessions, orders, conversion, sales, inventory, promotion status, and margin where supplied. Release the approved batch. Compare the agreed post-change window with the product's baseline and, where credible, similar unchanged products in the same store. Report the observed result, confounders, reversions, and confidence—whether the result is positive, neutral, or negative.
Want this control layer for your catalog?
The $495 founding Recovery Baseline covers up to 250 variants and three supplied exports. We confirm fit and the exact source list before requesting files or payment.
Request the Recovery Baseline