Skip to main content
Menu
Product
Solutions
Integrations
Resources
AI Catalog Mapping

Automatically match vendor columns to your product catalog

Every vendor names their columns differently — Item#, sku_code, PartNumber. AI Catalog Mapping reads the file, recognizes what each column actually is, and proposes a complete map to your clean schema. You review it once, adjust anything, approve it — and that mapping is saved for every future file from that vendor. AI suggests; you stay in control.

How mapping works

  1. AI reads the file Column names and sample values are analyzed together — 'blu wdgt 9.99' next to a header of 'desc2' still maps correctly.
  2. You review and approve Every suggestion is shown for review before anything publishes. Change a match, split a field, or leave a column unmapped.
  3. The mapping is saved Approved mappings lock in per source. The next file from that vendor maps itself — no re-review unless columns change.
  4. Rules refine every import Mapping rules and import rules apply your defaults, overrides, transformations, and exclusions automatically, file after file.
AI Mapping — in motion
Illustrative
SKU_00123 blue widget $9.99 in stock AI Mapping engine SKU SKU-00123 Name Blue Widget Price 9.99 USD Stock In stock
Illustrative diagram — raw vendor columns mapped to a clean schema

Why teams use it

Hours of column-matching, gone

The first map takes minutes to review instead of an afternoon to build. Repeat imports take zero.

Human approval, always

AI never publishes on its own. Suggestions wait for your review, and confidence is easy to judge with real sample values shown inline.

Your data standards, enforced

Replace values, normalize units, set defaults, and exclude rows with rules that run on every import — your catalog stays consistent no matter what arrives.

Survives format changes

When a vendor renames or reorders columns, changed columns are detected and re-proposed — known fields keep their mapping.

Cleanup built in

Generate consistent, channel-ready product names from messy vendor fields, fix typos, and restore broken SKUs as part of the same pass.

Reusable across catalogs

Mappings and rules belong to the source file, so every catalog built from it inherits the same clean result.

What's included

AI field recognitionSample-value analysisHuman review & approvalSaved mappings per sourceMapping rulesValue transformationsUnit normalizationDefaults & overridesRow exclusionsValidationAI product-name cleanupAutomatic reuse on future files

Good to know

What the AI does and doesn't do

It proposes mappings and cleanups from your file's structure and contents. It doesn't invent product data, and nothing it suggests becomes part of your published catalog until you've approved it.

Smart about cost too

Known sources reuse their saved mapping instead of re-running AI — so repeat imports are instant and your usage isn't spent re-answering the same question.

Works together with

Common questions

Is the AI mapping always right?

No AI is — which is why every mapping goes through your review before it's used. In practice most columns map correctly on the first pass, and your corrections are remembered so the same fix never needs making twice.

What happens when my vendor changes their file layout?

Changed columns are detected automatically and re-proposed for review; unchanged fields keep their existing mapping. Your rules and transformations carry over untouched.

Can I map the same file differently for different catalogs?

The field mapping belongs to the source file, keeping data consistent — while Catalog Rules control how each published catalog (and each reseller) sees products and pricing built from it.

Build one catalog that works everywhere.

Start a 14-day free trial and see it working on your own catalog — onboarding call included.