Migrating by hand: months of spreadsheets, errors and pain
Export to CSV, open Excel, map by eye, import, watch it fail, repeat. Wasted weeks and corrupted data.
The real risks
What’s at stake in a poorly planned migration
A migration without a methodology isn’t a system change: it’s an experiment with your operations, your accounts and your reputation.
Dirty data: duplicates, malformed IBANs, odd dates
Without cleaning it first, you’re migrating garbage. And that garbage contaminates the target system from day one.
No plan B: if something goes wrong, everything stops
Without a rehearsed rollback, a production problem stays unsolved and the business grinds to a halt until someone patches it up.
History that can’t be migrated as-is
Five years of delivery notes in an old schema, different accounting models, historical currencies. You need rules, not copy-paste.
Before vs after
Improvised migration vs. migration with a methodology
The difference between crossing your fingers on Monday and signing off the cutover with validated data, green tests and a rehearsed rollback.
Improvised migration
- Exports and CSV files edited by hand
- Mappings in Excel with no validation
- No automated tests
- Cutover on a Monday, “let’s see what happens”
- No rehearsed rollback
- A team crossing its fingers
Migration with a methodology
- Versioned, reproducible ETL
- Mappings with automatic validation
- Pre/post tests for every entity
- Cutover rehearsed several times beforehand
- Documented and tested rollback
- 2 weeks of hypercare after go-live
Methodology
Five phases designed to minimize risk
Each phase has verifiable deliverables and objective criteria for moving on to the next. Nothing is left to a last-minute judgment call.
1. Analysis
Inventory of data sources, data quality, risks and dependencies.
2. Mapping
Schema mapping with documented transformation rules.
3. ETL
Extraction, cleanup, deduplication and loading, reproducible and versioned.
4. Validation
Automated tests per entity: record counts, totals, referential integrity.
5. Cutover + Hypercare
Cutover and rollback rehearsals. Ongoing support for the 2 weeks that follow.
What we migrate
Types of migration we cover
Legacy systems, cloud platforms, ecommerce, CRMs and custom databases. If it holds data, we’ll move it without losing a single record.
Migration from legacy ERPs
Sage 50/200, ContaPlus, A3, ContaSol (popular Spanish accounting packages) and custom systems that have been piling up technical debt for years.
- Legacy schemas and historical currencies
- Chart of accounts and accounting model mapping
- Full history preserved
Migration between ecommerce platforms
Magento → WooCommerce, PrestaShop → Shopify, custom → headless. Catalog, orders and customers, with your SEO intact.
- Catalog, variants and attributes
- 301 redirects to preserve SEO
- Order history and reviews
Migration between CRMs
Salesforce, HubSpot, Zoho, Pipedrive and custom systems. Contacts, opportunities and activity brought together in one place.
- Contact and account deduplication
- Matching pipelines and stages
- Activity and interaction history
Database migration
MySQL, PostgreSQL, SQL Server, Oracle, MongoDB. Engine changes, normalization and schema consolidation.
- Type and constraint conversion
- Schema restructuring
- Verified referential integrity
File and document migration
DMS, document management systems, network repositories. Structure, metadata and permissions preserved.
- Folder structure and taxonomies
- Metadata and versioning
- Permissions by role and user
Cloud-to-cloud migration
AWS, GCP, Azure, OVH. Migration between providers or from on-premise without stopping your operations.
- Replication of services and configuration
- Delta sync right up to cutover
- Decommissioning plan for the source environment
Zero data loss guarantee
If something doesn’t fit, we roll back in under an hour
We don’t sign off go-live without a rehearsed, validated rollback. While your team checks the data in the new system, we keep the source environment intact and ready to switch back on. Business continuity never depends on luck.
Encrypted backups before and after cutover
Mirror staging environment for repeated rehearsals
Business validation window before the source is shut down
Documented rollback plan signed off by everyone
Automated post-load integrity tests
On-site hypercare for the 2 weeks after go-live
FAQ
Common questions about data migration
Which systems have you migrated?
Sage 50/200, Holded, FacturaScripts, Dolibarr, Odoo, ContaPlus, A3, ContaSol (popular Spanish accounting packages), CRMs (Salesforce, HubSpot, Zoho), ecommerce (Magento → WooCommerce, PrestaShop → Shopify), custom databases and spreadsheets.
How long does a typical migration take?
A standard ERP migration (customers + products + orders + invoices + accounting): 6–12 weeks, depending on data quality and volume. The longest part is usually the data cleanup.
How much does it cost?
The price depends on the data volume, the state of the data, the source systems and the validation required. Once we’ve reviewed them, we give you a proposal with scope, timeline and a fixed price.
Do you keep historical data?
Yes. Either by migrating the full history to the new system, or by keeping the old system in read-only mode for lookups. We choose based on volume and how often you need to consult it.
What happens if something goes wrong during cutover?
Every cutover is rehearsed in staging environments until we’re confident. If something doesn’t fit in production, we run the rehearsed rollback in under 1 hour and go back to the previous system.
Got a migration coming up?
We start with a free audit of your data and a risk map for the project. No obligation, no surprises: you’ll know exactly what you’re facing before a single record is moved.