How do you match legacy ERP records when there is no common identifier?

We compare combinations of available attributes, such as names, addresses, registration numbers and contact details. Different fields can use different comparison methods, with alternative rules for records missing particular attributes. The resulting matches need testing against known examples before consolidation.

Can you deduplicate ERP data without changing the source systems?

Yes. With WinPure, we can work on imported copies, investigate matches and prepare consolidated outputs while retaining the original source data. Writing results back is a separate, configured action. Your team can also export a mapping of source records to resolved entities for use in the migration.

Does a Golden ID replace the primary key in the new ERP?

A Golden ID provides a persistent reference for the resolved entity. Your ERP may generate its own primary keys, so the migration needs to preserve the mapping between legacy IDs, Golden IDs and target IDs. That mapping supports traceability and the reassignment of dependent records to the correct master.

What happens to invoices and orders when duplicate master records are merged?

The references attached to the original masters need to be mapped to the appropriate surviving record. Consolidating a supplier does not automatically update every invoice or purchase order in your ERP. We retain the source references needed to support that work; your migration tooling applies the mappings when loading the dependent data.

How should new and updated records be handled after the initial cleanup?

Subsequent extracts need to distinguish new entities from changes to existing ones. We can reuse saved cleaning and matching configurations and compare incoming records against established entities. The extraction design must also capture relevant updates and deletions, since a newly created record and a changed address require different treatment.

Can fuzzy matching resolve records with missing identifiers?

It can help where other populated attributes provide sufficient evidence. However, a similarity algorithm cannot infer a missing identifier simply by lowering the threshold. We assess which remaining fields can support the comparison and keep uncertain results separate for investigation.

Where does WinPure fit alongside an ERP migration or ETL tool?

We support the profiling, cleaning, matching, entity resolution and consolidation of the data being prepared. Your migration or ETL tooling handles the destination specific loading, including target mappings, transaction dependencies and ERP validation. The prepared records and identity mappings connect those two parts of the work.

Data Matching

Practical guidance on data cleansing, matching, and identity resolution - from the team building the tools.

All Product Updates Data Quality Data Matching Data Cleansing CRM Data Entity Resolution Data Match Algorithms

How to Manage Data Quality in an ERP Migration

In a recent ERP migration project, our team was knee-deep in helping a customer retain their legacy records that had been accumulated...

By Farah Kim, David Leivesley 23rd October 2026

Caverphone 2: How It Works, Examples & Free Calculator

What Is Caverphone 2? Caverphone 2 is a phonetic matching algorithm that converts text into a code representing an approximation of its...

By Team WinPure 18th October 2026

Introducing WinPure MatchGraph™: See the Relationships Behind Your Data Matches

Discover why records match, which rules connect them and where potentially over-merged groups deserve closer investigation. With WinPure Clean & Match Enterprise...

By Team WinPure 11th October 2026

Senzing Entity Resolution with WinPure: A Complete Guide to AI-Powered Entity Matching

Senzing entity resolution enables organisations to identify matching identities, detect duplicate records and discover relationships across fragmented datasets. As a purpose-built AI-powered...

By Team WinPure 11th October 2026

Introducing AdaptiveMatch™: The Next Generation of High-Performance Data Matching

WinPure Clean & Match Enterprise v12 marks a major leap forward in enterprise data matching technology, introducing a fundamentally re-engineered, high-performance matching...

By Team WinPure 11th October 2026

Build vs Buy Data Matching Software: What Does It Really Cost?

Thinking of Building Your Own Data Matching Software? Use our FREE Data Matching Software Build Cost Calculator to estimate the development effort,...

By David Leivesley 14th September 2026

Daitch-Mokotoff Soundex: How It Works, Examples & Free Calculator

What Is Daitch-Mokotoff Soundex? Daitch-Mokotoff Soundex is a phonetic matching algorithm that converts names into six-digit codes based on their likely pronunciation....

By Team WinPure 12th September 2026

Chapman Length Deviation: How It Works, Examples & Free Calculator

What Is Chapman Length Deviation? Chapman Length Deviation is a string-length similarity measure. It compares how long two text values are, without...

By Team WinPure 12th September 2026

Smith–Waterman–Gotoh Algorithm: How It Works, Examples & Calculator

What Is the Smith–Waterman–Gotoh Algorithm? The Smith–Waterman–Gotoh (SWG) algorithm is a string similarity algorithm that uses local sequence alignment to identify the...

By Team WinPure 12th September 2026