Simplify Data Logo
  • Services
      • App Development
      • Web Development
          • Wordpress Development
          • Wordpress Support
          • Accessibility
      • Software Development
      • IT Consulting SME
          • Legacy modernisation
      • Business Analytics
          • Power BI
          • Metabase
      • Artificial Intelligence
      • SEO & AI visibility
      • Maintenance & Support
  • Managed Services
      • Zero-Trust Networking VPN
      • Managed email & groupware
  • Products
      • QRFlow
      • WooCommerce Interface
      • Open Source Projects
  • About us
      • Career
  • Blog
  • Contact
IT Consulting

Data Migration in Legacy Modernisation: How to Avoid Losing Data

Simplify Data Team
November 16, 2026
5 Min. Read
Data Migration in Legacy Modernisation: How to Avoid Losing Data
Simplify Data
Simplify Data GmbH
www.simplify-data.de
Data Migration in Legacy Modernisation: How to Avoid Losing Data

Simplify Data Team • November 16, 2026 • 5 min Lesezeit

When you modernise an old program, you tend to think of the interface and features first. The real risk usually lies elsewhere: in the data. Over the years, customer records, orders, articles and histories have grown, often with their own rules, special cases and legacy baggage. If they get lost or distorted during the switch, even the nicest new application is of no use. This article shows how a data migration runs in a controlled way. The general course of a modernisation is covered in our article "Modernising Legacy Software: Step by Step, Not a Rewrite"; here we focus on the data.

Why the data is the real risk

An application can be rebuilt, your business data cannot. It is the result of years of work and is often stored in the old application in ways that were never documented: status values with special meanings, free-text fields into which information was entered for other purposes, or calculations that exist only in the program code. That is why we treat data migration as a separate workstream with planning, testing and acceptance.

Step 1: Take stock of your data

Where is the data really?

It starts with an inventory. Besides the main database there are often further sources: Excel lists on network drives, individual Access files, exports for accounting or files that employees maintain locally. We record data sources, formats, volumes and dependencies on other systems.

What is still needed?

Not every record has to move. Together with the departments we clarify which data is needed in daily business, which only has to be archived and which can be dropped. This makes the migration smaller and reduces risk.

Step 2: Clean the data

Legacy data almost always contains duplicates, incomplete entries, inconsistent spellings and orphaned links. If you take it over uncleaned, you carry the problems into the new system. Typical tasks are:

  • Detect and merge duplicates, for example customers created twice
  • Check mandatory fields and fill in missing details
  • Standardise formats, for example dates, phone numbers or addresses
  • Identify orphaned records without any reference and assess them together

The business decision of what applies is made by your team. We provide analyses that make this decision easier.

Step 3: Map the old data model to the new one

The new system usually has a clean data model that differs from the old one. In the mapping we define which old field flows into which new field, which values are converted and where data is split or merged. This mapping is documented and agreed with the business owners. That way it stays traceable why a value looks the way it does in the new system. We also work out the target system and its technology stack together with you, because the data model depends on it.

Step 4: Trial migrations and reconciliation

Before real data moves, we practise. In trial migrations the data is loaded into a test environment and checked against the existing data.

Counts, totals, samples

Three kinds of checks have proven their worth:

Check What it shows
Count Does the number of records per table or module match?
Totals Do amounts, quantities or balances match between old and new?
Sample Do selected records, including special cases, look correct from a business point of view?

Deviations are analysed, the mapping is adjusted and the trial migration is repeated. Only when the checks pass consistently is the data ready to be taken over.

Step 5: Migrate iteratively, module by module

The data does not move all at once but module by module, matching the iterations of the modernisation. First comes the customer base, for example, later orders or invoices follow. Each stage is checked and accepted by you before the next one begins. This keeps risk manageable and delivers visible results early.

Parallel operation and rollback

As long as not all modules have moved, old and new run side by side. This needs clear rules: which system is the leading one for which data, and how are changes reconciled? We also plan a rollback in advance. We back up the state before every migration and define under which conditions we return to the old system. A rollback plan that is never needed is cheaper than an unplanned outage.

Retention and data protection

A migration often moves personal data and documents with retention periods. In general, data must not end up uncontrolled in test environments or on private devices, and retention periods must also be observable in the new system. Which periods and obligations apply in your case should be clarified with your tax adviser or your data protection officer. We take care of the technical implementation.

Typical risks and countermeasures

Risk Countermeasure
Data is lost or truncated Counts and totals for each trial migration
Errors from legacy data move into the new system Cleaning before the migration
Unclear meaning of fields Agree the mapping with the department and document it
Switch fails on the cut-over date Module-by-module migration and parallel operation
No way back Backup before every step, defined rollback

Checklist for your data migration

  1. Record all data sources, including Excel lists and local files.
  2. Decide with the departments what is migrated, archived or deleted.
  3. Clean duplicates and legacy baggage before the transfer.
  4. Document the field mapping and have it accepted by the business.
  5. Run trial migrations with counts, totals and samples.
  6. Migrate module by module and accept each stage.
  7. Plan parallel operation with clear rules on which system leads.
  8. Define a backup and a rollback plan before every step.
  9. Have retention periods and data protection reviewed by specialists.

How Simplify Data can help

We accompany your legacy modernisation from the assessment to the data transfer. We work out the target stack together with you, and the migration runs iteratively with acceptance after every step. You can find more about our approach on the page Modernising legacy systems. If you have an old application with a grown data set, get in touch.

This article is for general information only and is not legal or tax advice; for your specific case please contact a lawyer or your tax adviser.

© 2026 Simplify Data GmbH www.simplify-data.de
Back to Blog

Table of Contents

  • 1.Why the data is the real risk
  • 2.Step 1: Take stock of your data
  • –Where is the data really?
  • –What is still needed?
  • 3.Step 2: Clean the data
  • 4.Step 3: Map the old data model to the new one
  • 5.Step 4: Trial migrations and reconciliation
  • –Counts, totals, samples
  • 6.Step 5: Migrate iteratively, module by module
  • 7.Parallel operation and rollback
  • 8.Retention and data protection
  • 9.Typical risks and countermeasures
  • 10.Checklist for your data migration
  • 11.How Simplify Data can help

Actions

Our services
  • App Development
  • Web Development
  • Software Development
  • IT Consulting SME
  • Business Analytics
  • Artificial Intelligence
  • SEO & AI visibility
  • Maintenance & Support
  • IT services Kiel & Hamburg
Service & Legal
  • Legal notice
  • Privacy policy
Contact
  • Simplify Data GmbH
  • Königsweg 48, 24114 Kiel
  • Phone: 0431 705 770 24
  • Email: info@simplify-data.de

Copyright © 2026 Simplify Data GmbH