Most platform consolidation conversations start with cost. Snowflake licensing is expensive. Maintaining a third-party data lake alongside a Microsoft Fabric investment creates overlap most organizations can no longer justify. The business case for migration is usually straightforward.
What organizations don’t anticipate is what happens when Informatica’s Intelligent Data Management Cloud (IDMC) starts profiling the data before the move.
Migration is framed as a logistics problem. Get the data from point A to point B. Validate completeness. Cut over. But when IDMC’s data quality and lineage capabilities are applied to the source environment in preparation for migration, something else happens: the organization gets a detailed picture of what’s actually in the lake – and in most cases, it’s not what anyone expected.
That picture is as valuable as anything waiting on the destination side. And most organizations aren’t prepared to use it.
What the Data Actually Looks Like Before the Move
Third-party data lakes accumulate over time. Data gets loaded for a use case that no longer exists. Schemas get duplicated across teams that couldn’t agree on a shared definition. Quality exceptions get handled downstream, in the reports or in the analysts’ heads, rather than at the source. None of this shows up as a problem until something tries to use the data systematically.
IDMC’s profiling and discovery capabilities surface the real picture quickly. What we consistently find when working with clients migrating off Snowflake:
- Duplicate records that were masked by downstream reconciliation logic
- Datasets that haven’t had a consumer in months or years, still being stored and paid for
- Column definitions that exist only in tribal knowledge – no documentation, no governance, no way to validate
- Lineage gaps where no one can trace an output back to a trusted source
- Quality failures that reporting layers have been quietly compensating for
None of this is unusual. It’s the natural result of a platform that grew organically over time without a governed data layer underneath it. The problem isn’t that the organization did something wrong. The problem is that migration forces a choice: move what’s there, or move what’s worth keeping.
Migration as an Unplanned Audit
This is the reframe most organizations miss. The value of IDMC in a migration context isn’t just that it moves data reliably. It’s that it forces a conversation about data quality, lineage, and ownership that the business has been deferring.
Every organization has that conversation eventually. The ones that have it during migration – when there’s a forcing function and a destination to build toward – are in a better position than the ones that carry unresolved problems into Fabric and discover them later, at scale, when the cost of fixing them is higher.
The table below captures what IDMC typically surfaces during migration profiling and what to do with each finding before the data moves:
| What IDMC Finds | What it Actually Means | What to Do Before You Move |
| Duplicate or conflicting records across schemas | The same entity has been defined differently in different parts of the lake | Master the entity before migrating – don’t carry duplicates forward |
| Orphaned datasets with no downstream consumers | Data that was loaded and never used – still paying for storage and compute | Decommission before migration; don’t replicate cost without value |
| Columns with undocumented definitions | Business rules living in analyst memory, not in the platform | Document and govern definitions before they become Fabric’s problem |
| Data lineage gaps – source unknown | Outputs that can’t be traced to trusted input | Resolve lineage or flag for review; unknown provenance doesn’t improve in transit |
| Quality failures that reports have been working around | Downstream consumers compensating for bad data upstream | Fix at the source, not the workaround layer |
What This Means for the Fabric Environment
The destination matters. But what arrives at the destination matters more.
Fabric is a capable platform. IDMC connects to it two ways – MDM Extensions bring mastered data in as a foundation, and data quality runs once it’s there. But neither the platform nor the integration layer can compensate for data that arrives unmastered, undocumented, or untraced. Fabric inherits whatever is migrated into it – the good and the unresolved.
Organizations that use the migration as an opportunity to govern the data in transit arrive at Fabric with something they didn’t have before: a cleaner foundation, documented lineage, and a starting point for the master data work that Power BI and downstream AI capabilities will depend on.
The ones that treat migration as a logistics exercise arrive at Fabric faster – and spend the next twelve months dealing with the same data quality problems in a new environment.
The Cost Consolidation Case Is Real – But It’s Not the Whole Story
Eliminating a Snowflake contract – or any third-party data lake – in favor of Fabric is a legitimate cost play. Compute, storage, licensing, and integration overhead all compress when the platform footprint consolidates. For organizations already invested in the Microsoft ecosystem, the consolidation case often writes itself.
But the organizations getting the most out of consolidation aren’t just reducing spend. They’re using the migration as the moment to resolve the data problems that were always present but never prioritized. The cost savings fund the governance work. The governance work makes the new platform worth what was spent on it.
IDMC is what connects those two outcomes. Its quality, lineage, and profiling capabilities in transit turn a cost decision into a data strategy decision. And that’s a different kind of return.
The Bottom Line
Platform consolidation off a platform like Snowflake onto Microsoft Fabric makes financial sense for most organizations already in the Microsoft ecosystem. But the migration is also the best opportunity most organizations will have to understand what their data actually looks like – before it becomes Fabric’s problem.
IDMC’s quality and lineage capabilities during transit aren’t a migration feature. They’re a data strategy accelerator. The organizations that treat the move as an audit – and act on what they find – arrive at Fabric with a foundation worth building on.
By Tom Rudnick, Senior Solutions Executive and Nicklaus Porter, Senior Data Analyst