Reconcile
Join SKUs, variants, source records, assets, and channel listings. Persist confirmed identity and isolate ambiguity.
Catalog recovery research for multichannel brands
Find catalog records that deserve attention. Fix them safely. Measure what changed.
We’re testing whether growing brands need a better way to reconcile product truth across Shopify, Faire, Google, source files, and performance exports—without adopting another platform.
Research stage only. No purchase, files, credentials, or promised pilot seat.
VERIFIED EXPOSURE MAPSYNTHETIC EXAMPLE
$26.9Kpositive historical net sales on high-priority products
11products
2channel blocks
2products with identity exceptions
4fix first
GUEST-PNK-01 LIMITEDCAN-AMBER-8 HIGH NET SALESChange brief CONTROLLED
Cross-source identityVersioned channel rulesControlled changes + outcomes
The gap between tools
Your software can edit. It does not own the handoff.
AI writers can generate copy. Bulk tools can move rows. PIMs can centralize data. The expensive work remains: deciding which record is true, which defects matter commercially, which changes are safe, and whether the release improved anything.
Catalog Asset Care is exploring a managed recovery loop that works with the merchant's existing stack instead of asking the merchant to operate another platform.
One bounded recovery loop
Reconcile. Prioritize. Control. Prove.
Prioritize
Combine verified defects with sales, traffic, inventory, eligibility, confidence, and effort so high-value work rises first.
Control
Prepare evidence-backed diffs, validate dependencies, route exceptions for approval, and package backup and rollback.
Prove
Verify the release and compare the agreed baseline and post-change window with attribution limits stated plainly.
What makes the paid work different
One evidence chain, two controlled stages.
The $495 Baseline establishes product truth and the recovery decision. Only after a successful Baseline may a $1,500 Sprint prepare approved changes, a release pack, verification, and measurement.
$495 Recovery Baseline · decision package
Product identity graph
Canonical products, variants, SKUs, assets, and channel records joined with confidence and a bounded exception queue.
Verified exposure map
Every defect tied to its evidence, source, affected channel, severity, and the historical revenue or inventory attached to it.
Commercial recovery queue
Work ranked by actual sales, traffic, stock, eligibility impact, repair confidence, and effort—not by a generic content score.
Source plan + controlled change brief
Source precedence, unresolved decisions, destination-rule evidence, and bounded next actions—without changing or publishing catalog records.
$1,500 Verified Recovery Sprint · only after a successful Baseline
Approved-fact fix pack
Proposed titles, attributes, descriptions, and asset actions grounded only in customer-approved facts, with unsupported claims blocked.
Safe release pack
Before/after diffs, validation, customer approval, import-ready files, exception report, backup, and rollback instructions.
Recovery measurement ledger
A durable record of approved changes, releases, reversions, and the post-change measurement window.
Best fit
Growing catalogs that have outgrown a founder's spreadsheet—but do not want a PIM project.
Good fit
- 200–5,000 active SKUs or variants
- Shopify plus Faire, Google, or another destination
- Three or more product-data or asset sources
- Seasonal launches, supplier updates, or recurring feed drift
- Existing sales, traffic, or wholesale demand
Not a fit
- One clean channel and fewer than 100 SKUs
- Only generic AI descriptions are needed
- Real-time ERP/PIM integration is required immediately
- Automatic publishing without product-owner approval
- Regulated or safety-critical claims in the founding pilot
Proposed model · currently in research
See the risk free. If the need is real, establish truth for $495.
These are the controlled offers being tested—not services currently available to purchase. Payment, confidential files, and delivery remain closed while professional-readiness work continues. Interest-list evidence will determine whether a small pilot should open at all.
Recovery Preview
$0Up to 10,000 CSV rowsLocal catalog scan, optional performance and channel-issue join, observed revenue exposure, and downloadable recovery queue.
Run it now →Recovery Baseline
$495Founding offer · up to 250 variantsThree supplied exports, verified identity and rule review, top-ten recovery queue, source plan, and controlled change brief.
Join interest list →Verified Recovery Sprint
$1,500Founding offer · top 25 opportunitiesApproved-fact fixes, evidence ledger, import-ready pack, backup and rollback plan, verification, and 30-day readout.
Recovery Assurance
$599/moAfter a successful sprintMonthly delta scan, up to 50 changed records, bounded exception resolution, channel drift check, and outcome report.
Joining the interest list does not reserve a place, create a contract, or authorize Catalog Asset Care to request files or payment.
Why this is not a weekend app
The interface is the easy part. Maintaining trust is the product.
Could we build a CSV checker with ChatGPT?
Yes—and that is why the free preview is not the paid product. The paid system combines multiple sources, persistent identity decisions, maintained channel rules, human exceptions, change-risk controls, and measured outcomes.
Why not use Matrixify, an AI editor, or a PIM?
Use them where they fit. Matrixify is excellent transport; AI editors are efficient drafting tools; a PIM is appropriate for a staffed data program. We operate the cross-tool control loop and can hand approved changes to those systems.
Where does GenAI fit?
Downstream of verified facts. It can draft channel-specific fields and explain issues, but every factual claim keeps its source and approval state. Unsupported claims are blocked instead of invented.
Do you claim a precise amount of lost revenue?
No. We separate observed breakage, historical revenue exposure, modeled scenarios, and measured post-change results. Assumptions and confidence remain visible.
Do you publish changes?
Not during the founding pilot. We deliver a validated, minimal change set, backup, and rollback instructions. The customer approves and publishes through its existing process.
Will we need to operate another app?
No. Customers provide bounded exports, answer only true exceptions, approve the change set, and receive the release and outcome packs. The internal workflow is operated for them.
Founding Pilot Interest List
Tell us whether this problem is real before we build a service around it.
Join the research list if catalog identity, asset matching, feed drift, or risky bulk changes create meaningful work for your team. We’ll use aggregate interest to decide whether to open a small pilot.
No commitmentNo seat promised · no purchase · no files or credentials · optional research conversation · contact permission required