Checklist

Data migration checklist for ERP and CRM projects

Every ERP or CRM project stands on the data it inherits. Item lists with four spellings of the same product, customers duplicated across sheets, balances that do not tie to the accounts — imported as they are, they undermine every report the new system produces. This checklist covers the migration from first inventory of sources to the day the old spreadsheets are made read-only.

Why migration deserves its own plan

Data migration is the least visible part of a system project and the one most likely to delay go-live. It touches every department, depends on people who are busy with their day jobs, and exposes years of accumulated inconsistency. Treated as an afterthought, it becomes the reason the new system is not trusted; treated as its own workstream with an owner and a timeline, it becomes the foundation of every report that follows.

The checklist below runs in order: inventory, clean, map, rehearse, cut over, retire. Each stage should be visibly complete before the next begins.

How to use this checklist

Assign each item to a named person with a date. Keep a log of every cleaning rule and mapping decision — it will be needed when a total does not tie or a colleague asks why a customer was merged. Do not tick 'reconciled' until the numbers agree on paper, signed off by whoever owns the accounts.

  • Every source of data is inventoried — spreadsheets, paper registers, accounting software, old systems, personal laptops and drives
  • For each source: what it holds, who owns it, who edits it, how often, and which other files it feeds
  • Formulas, macros and manual rituals hidden in spreadsheets are documented as business rules to reproduce or replace
  • One named person owns each master data set — items/SKUs, customers, suppliers, warehouses, chart of accounts
  • Naming conventions are agreed for items, customers and suppliers before cleaning starts
  • Duplicates are merged and dead records archived rather than imported into the active system
  • Units of measure, date formats, phone formats and codes are standardised across sources
  • Information hiding in notes fields (sizes, colours, payment terms) is moved into structured fields where the new system needs it
  • Opening balances — stock per location, receivables and payables per party — are reconciled to the accounts before import
  • A field-mapping table exists (source column → target field) and has been reviewed by the business owners, not only the technical team
  • The decision on history is written down — master data and open transactions always; closed history summarised or selectively imported
  • Cleaning is done in a copy of the data with a log of every rule applied
  • The import has been rehearsed at least once on a test environment with the full data set
  • Totals are compared between source and target — item counts, stock value, receivables, payables — and differences resolved
  • A cut-over moment with low activity is chosen; spreadsheets are frozen at that point and the final import runs
  • Old and new systems run in parallel for a short, defined period with daily comparison of key figures
  • Retired spreadsheets are made read-only, moved to an archive folder and the change is announced to the team
  • Backups of both the original files and the migrated database are taken and a restore has been tested
  • Data-protection requirements for personal data (customers, employees) are respected during migration and archiving

Frequently asked questions

Usually longer than the owner expects and often as long as the build itself. Cleaning years of registers, standardising item names, de-duplicating customers and reconciling opening balances is careful work. Start it in parallel with design, not after the software is ready.

Migrating data into a new ERP or CRM?

Book a free consultation. We will review your current files, estimate the cleaning effort honestly and show you how a rehearsed, reconciled import fits into a phased go-live.