# Data migration and clean-up

> Data migration moves records — customers, deals, products, history — from an old system to a new one. DevRaxo maps fields, cleans and de-duplicates the data, runs test migrations, and cuts over with a rollback plan, so the new system starts accurate and the old one can be retired without losing history.

- Area: [CRM & integrations](https://devraxo.com/services/crm/)
- Price: from £1,200. Larger or multi-system migrations are scoped after the audit.
- Timeline: 2–4 weeks
- Page: https://devraxo.com/services/crm/data-migration/

## This is for you if
- You are replacing your CRM, shop, or an old in-house system.
- Years of data are duplicated, inconsistent or half-empty.
- You are nervous that a switch will lose customer history.

## The problem
- **Garbage in, garbage forever.** Importing messy data into a clean new system just moves the mess somewhere more expensive.
- **Fields do not match.** Old and new systems describe a customer differently, so naive imports lose meaning.
- **There is no way back.** A one-shot import with no rollback turns a mistake into a crisis.

## How it works
1. **Audit what you have.** Volumes, quality, duplicates and the fields that actually matter, with an honest estimate of the clean-up.
2. **Design the mapping.** Every field from the old system mapped to the new, with rules for the awkward cases.
3. **Test migrations until they are boring.** Repeated trial runs into a test system, checked against the source, until the numbers match.
4. **Cut over with a rollback plan.** A planned switch-over window, verification checks and a tested way back if anything is wrong.

## What you get
- Data audit and clean-up report
- Field mapping specification
- Test migrations with reconciliation
- Production cut-over
- Rollback plan
- Archive of the source data

## Questions
### Will we lose any history?
Not if it is mapped properly. We reconcile record counts and sample records between old and new, and keep a full archive of the source data.

### Can you migrate from a system that is being shut down?
Yes, often urgently. We start by securing a complete export so the timeline pressure does not put the data at risk.

### How do you handle duplicates?
We use matching rules you approve, merge the obvious ones automatically, and list the uncertain ones for a person to decide.

### Do we have to stop working during the cut-over?
Usually for a short, planned window — often an evening or weekend. We agree it in advance and keep it as brief as possible.