Ask better questions before data migration

Define success before moving data

Migration acceptance should be understandable before cutover. Identify the datasets that matter, the business owners who can confirm them and the checks that support acceptance. ## Document the transformation Source and target systems may classify or store information differently. Record the agreed mapping and transformation rules. Identify what happens to rejected or incomplete records and who decides how exceptions are handled. ## Plan reconciliations A reconciliation should explain what is compared and why that comparison matters. Counts, totals and selected field checks can answer different questions. The approach depends on the data and the engagement scope. ## Preserve traceability Keep records of the migration runs, exceptions, decisions and validation performed. Reviewers need to understand which dataset and run produced the evidence they are examining. ## Acceptance questions - Who owns each critical dataset? - Are mapping and transformation rules approved? - Which checks support completeness and accuracy? - How are exceptions investigated? - Who approves acceptance or rollback decisions? ## Next step A data migration audit can review the agreed controls and evidence. It supports management decisions without guaranteeing that every error has been identified.