Data warehouse and analytics consolidation: Snowflake, BigQuery and Databricks
The warehouse is rarely the hard part. Agreeing what a customer, an order and revenue mean in both companies is.
Settle metric definitions and customer identity first, then choose a pattern: federate (query across both without copying), replicate one into the other, or migrate to one platform. Keep both warehouses running until the shared definitions are tested, and re-point dashboards last.
Snowflake
Google BigQuery
Databricks
Amazon Redshift
Fivetran
dbt
How to read this chart
The scores are our planning judgement for a typical small or mid-size deal: they show where effort tends to concentrate, not measured data. Replace them with your own once the inventory is done.
What should be agreed before moving any data?
A short data dictionary for the metrics leadership uses: revenue, customer, active user, churn and margin. Without it, the combined dashboard shows two truths.
- Name one owner per metric.
- Write the definition and the source table for each side.
- Decide how the two definitions will be reconciled or reported separately.
How do you match customers across two companies?
Use a stable key such as domain or email, review near matches, and store the result in a mapping table rather than overwriting source data.
- Normalise names, domains and emails on both sides.
- Match and score; accept clear matches automatically.
- Review the uncertain cases with the account owners.
- Publish the mapping table for reports and the CRM merge to share.
What about access and privacy?
Merging data also merges who can see it. Rebuild roles from the combined org chart rather than copying each side's grants, and check what you are allowed to combine before you do.
- Review personal and payment data against the purposes it was collected for.
- Give analysts read access to what they need and no more.
- Log who queries sensitive tables.
Questions on this topic
Do we need one warehouse?
Not necessarily. Many combined companies run two for a long time and share a reporting layer. Consolidate when running both costs more than the move.
Which warehouse should win?
The one that fits the combined skills, workloads and contracts. Check each vendor's current documentation for sharing and migration options.
Where do pipeline tools fit?
They load data from source systems into the warehouse. Re-pointing them is usually part of the cutover, so inventory them in diligence.
Vendor documentation to check against
Product details change. Confirm current options in each vendor's own documentation. Consultants in a Box is an independent consultancy; this page is not published or endorsed by any vendor named. Names and logos are trademarks of their owners.
Related guides
CRM merge
A CRM merge is mostly a data-quality and ownership project. Merging records is the easy part; deciding what a customer is, who owns them and which fields survive is where the time goes.
Read the guideERP and accounting
Finance cannot pause. Keep both sets of books running through the first closes, agree one chart of accounts for reporting, and move to a shared system only when the close process is stable.
Read the guideSynergy capture
Tech savings are real but smaller and slower than deal models often assume. Count the one-time cost to capture them, and size them with your own contract numbers.
Read the guideAll M&A tech stack guides
The full playbook, from diligence to synergy capture.
See the hubMerging two stacks?
Start with the free business analysis, or book a call and we will talk through both companies' systems and where to begin.
