Solutions / Catalog migrationSchema crossing / migration bridge

Catalog migration

Do Not Migrate Bad Product Data into a New System

A platform migration changes where product information lives. It does not automatically improve the information itself.

If a legacy catalog contains inconsistent attributes, missing specifications and outdated structures, moving those records unchanged simply transfers the problem into a newer system.

ENRIVAQ positions enrichment and cleanup before migration so the target platform receives product data that is closer to the structure the business actually wants to manage.

Legacy catalog
Historic taxonomyLegacy spare-parts groupstructure inherited
Duplicate conventionOuter Ømapping required
Source record
As receivedLEGACY
Identifier
AC820825
Attributes
Limited source fields
Technical data
INCOMPLETE
Category
Legacy category
Source preserved · decisions not implied
Migration bridge
CleanupNormalizeEnrichMap
Pilot gateVALIDATEbefore scale
Target catalog
Reference product AC820825
Connected product identity

Target record

Kverneland AC820825 Fan Impeller
CUSTOMER-APPROVED MODEL
Canonical fields
OEM / MPN · EAN · product type · compatibility · accepted technical fields
Normalized units
Canonical values; unresolved weight conflict excluded
Target taxonomy
Target taxonomy
Review state
PILOT GATE
No unsupported value crosses the gate

LEGACY CATALOG → CLEANUP / NORMALIZE / ENRICH / MAP → PILOT GATE → TARGET CATALOG

Migration problem

Garbage in, garbage out

A new PIM or ecommerce platform can enforce better workflows, but only after the imported data has been mapped into a usable structure.

Migration is therefore a good moment to decide what the catalog should look like on the other side.

Skipping that step can make the new system inherit the same duplicate fields, inconsistent units and incomplete product records as the old one.

Migration mirror
Legacy record1:1 COPYNew platform
Attribute namessupplier-specific names→canonical names
Unitsmixed units / formats→mixed units / formats
Required valuesmissing or incomplete→missing or incomplete
Categorieslegacy hierarchy→legacy hierarchy
The interface changes. The record does not.

Audit current catalog

Create a migration baseline before transforming the data

The first step is to understand the source catalog. Review product categories, identifiers, attribute sets, missing fields, duplicate structures, data formats and current content.

The audit should also identify which parts of the catalog are representative enough for a pilot migration.

Product data cleanup →
Audit baseline board
01Product categoriesCHECK IN SAMPLE
02IdentifiersCHECK IN SAMPLE
03Attribute setsCHECK IN SAMPLE
04Missing fieldsCHECK IN SAMPLE
05Duplicate structuresCHECK IN SAMPLE
06Data formatsCHECK IN SAMPLE
07Current contentCHECK IN SAMPLE
08Representative pilotPROJECT-DEFINED SAMPLE
Baseline findings are documented from the supplied source export — no result is assumed.

Normalize

Standardize the information that already exists

Before mapping into a new structure, normalize values that represent the same concept differently. This may include units, attribute names, capitalization, formats and controlled values.

The target format should come from the new catalog model rather than from arbitrary AI preferences.

Product data standardization →
Normalization crossing
Legacy conventionsAPPROVED RULETarget model
Unitscm→mm
NamesOuter Ø→outer_diameter
Capitalizationsource capitalization→catalog convention
Controlled valuessupplier value→controlled value
Target model decides · customer-approved schema required

Enrich

Complete important information before it reaches the target system

Migration is an opportunity to improve records that have remained incomplete for years. Where reliable product information can be found, enrichment can add missing technical attributes, identifiers or content before the records are loaded.

Enrichment sieve
Gap in legacy recordEvidence testGate decision
Required technical attributeRelevant source supports the valueENRICH
Required identifierVerified identity relationship existsENRICH
Critical valueNo supporting evidence suppliedLEAVE BLANK
Optional legacy fieldNot required by target catalogSKIP
Not every blank field must be filled. Priority should be given to fields that matter for the target catalog and can be supported by evidence.

Map taxonomy

Map products into the structure the new catalog will use

Legacy categories often reflect the history of the old system rather than the product logic desired in the new one.

Taxonomy bridge
Source taxonomyTarget taxonomy · attribute family
Legacy group A→Mapped target categoryCategory attribute family
Legacy group B→Mapped target categoryCategory attribute family
Unresolved legacy category→REVIEW REQUIREDNO AUTO-MAP
Mapping products into the approved target taxonomy gives downstream enrichment, filters and content generation a consistent structure.

Validate

Test the transformed record before full migration

Normalization, enrichment and category mapping can all introduce errors if rules are wrong. Validation checks whether transformed records match the target schema, whether required fields remain missing and whether technical values or mappings need review.

A representative pilot should be validated before scaling the migration.

Representative pilot
Validation checkpointVERIFY BEFORE SCALE
01Target schema matchVERIFY IN PILOT
02Required-field statusVERIFY IN PILOT
03Technical values and mappingsVERIFY IN PILOT

Export to new system

Deliver data in the format required by the migration workflow

Once the transformed records are approved, they can be exported toward the target PIM, ERP-related flow or ecommerce platform through supported methods.

The exact import/export process depends on the systems involved and must reflect real integration capabilities.

Verified transfer gate
DestinationProject methodTransfer state
Target PIMConfigured · native connector not claimedPILOT FIRST
ERP-related flowAPI/file input · write-back configuredPILOT FIRST
Ecommerce platformConfigured site export where implementedPILOT FIRST
TRANSFER CONDITIONApproved record structure · tested format · supported write path

Migration checklist

A practical product-data checklist before go-live

Use the migration project to confirm that product information is ready for the new environment.

Go-live control board
01Target taxonomy approved.CHECK
02Required attributes defined by category.CHECK
03Legacy fields mapped or retired.CHECK
04Units and naming normalized.CHECK
05Priority missing information enriched where possible.CHECK
06Technical values validated.CHECK
07Exceptions reviewed.CHECK
08Export format tested on a representative batch.CHECK
09Rollback / source preservation handled by the migration project owner where required.PROJECT OWNER

Controlled next step

Use a migration sample to find data problems before go-live

Share a representative source export and the target catalog structure to assess the cleanup and enrichment work required before migration.

Talk to ENRIVAQ

Request a catalog assessment

Tell us enough to make the next step useful for your catalog.