Something is happening at Microsoft. I've been watching it since early 2019 and it's becoming harder to ignore. ADF and Power BI are converging. Not just in marketing language — in actual product architecture. Let me tell you what I'm seeing and what I think it means.
The Evidence
Shared Engine: Power Query in ADF
Wrangling Data Flows use the Power Query M engine. This isn't a coincidence or a convenience hack — it's an architectural decision to share the data preparation runtime across ADF and Power BI. The same engine that runs M formulas in Power BI's Get & Transform now runs M formulas in ADF pipelines.
Shared engines require coordination between product teams. The Power Query team is serving two product owners now. That's an organizational structure that reflects a strategic decision about platform convergence.
Shared UX Language
The ADF canvas authoring experience has been evolving toward the Power BI visual style. The node-based pipeline graph in ADF now looks more like a Power BI data model diagram than it did in 2017. Whether this is a coordinated UX convergence or parallel evolution toward similar design patterns, the result is a consistent visual language across the products.
Marketing Positioning
Microsoft product communications have started grouping ADF and Power BI consistently under "Azure Data and Analytics." The data stories at Microsoft conferences in 2019 present ADF, Power BI, Azure SQL, Azure Synapse (just announced), and Azure Databricks as a unified platform. ADF and Power BI are no longer presented as separate products with separate positioning — they're part of the same data platform narrative.
The Azure Synapse Announcement
In May 2019, Microsoft announced Azure Synapse Analytics (building on what was Azure SQL Data Warehouse). The preview positioning includes something called "Synapse Pipelines." The Synapse Pipelines screenshots look — very familiar. The activity types. The canvas. The linked services model. The pipeline JSON structure.
Synapse Pipelines is ADF, or something very close to it, embedded in the Synapse workspace.
If ADF pipelines are going to live inside Synapse Analytics — the platform Microsoft is positioning as the unified analytics workspace — then ADF as a standalone product might be in a transitional phase. The capabilities are real and being invested in. Where they'll live in three years is a different question.
Historical Pattern: Microsoft Product Convergence
Microsoft has done this before. SQL Server Reporting Services and Power BI overlapped for years — similar capability, different deployment models, different teams. Eventually Microsoft's investment clearly shifted toward Power BI and SSRS became the legacy path.
SSIS and ADF have been coexisting since 2014 in exactly this way. SSIS is still supported. Investment in SSIS features has slowed. The new thing — ADF — is where the product energy is.
Is ADF now in the SSIS position relative to Synapse? Possibly. The investment in ADF is clearly real — the 2019 feature releases are substantial. But the Synapse announcement signals that the feature investment may be shipping into Synapse Pipelines as the long-term home, with ADF as the current-generation interface.
What This Means for Teams Building on ADF Now
Here's the practical implication: the capabilities you're building with ADF are real and will exist. The patterns — parameterized pipelines, ForEach, Lookup-driven config, CI/CD via ARM templates — are sound and transferable. If the product home moves from ADF to Synapse Pipelines, the migration is likely to be a re-registration of the same pipelines into a new workspace, not a rebuild of the logic.
This is the Microsoft pattern: the capability moves, the surface area changes, the code mostly ports. Your investment in ADF patterns is an investment in the Azure data pipeline pattern, not specifically in a product brand that might be renamed.
The risk: if you've built tight integrations with ADF-specific tooling (custom ARM template scripts, ADF REST API wrappers, SHIR management automation) rather than against the pipeline patterns themselves, a product home change is more work. Build to the patterns, not the surface API.
The Near-Term Outlook
ADF is being actively invested in. The 2019 releases — Data Flows GA, Wrangling Data Flows, monitoring improvements, new connectors — are real. Microsoft is not winding ADF down. The Synapse announcement is about where the long-term unified platform lives, not about ADF going away in the next two or three years.
Watch Synapse Pipelines closely. If it ships with feature parity to ADF plus additional integration with Synapse's SQL and Spark capabilities, the migration path from ADF to Synapse Pipelines becomes the natural next step. Microsoft will make that migration easy — it's in their interest to move customers forward rather than strand them on an older product.
For now: build on ADF v2, invest in the patterns, watch Synapse. The tea leaves say convergence is coming. The timeline is 2020-2022, not 2019. You have time to build on solid ground. As always, I'm here to help if you want to talk through platform strategy for a specific client situation.