On October 13, 2025, Fivetran and dbt Labs announced an all-stock merger, creating a combined entity approaching $600M ARR under CEO George Fraser, with Tristan Handy as President. Their stated goal: “open data infrastructure” unifying data movement, transformation, metadata, and activation. Fivetran also acquired Census (reverse ETL, May 2025) and Tobiko Data/SQLMesh (September 2025).
The broader merger implications — the Core/Cloud divergence, the challenge to Dataform, the end of the pure best-of-breed era — are covered elsewhere. This note focuses on one specific consequence: what the merger means for orchestration choice.
The Orchestration Independence Argument
The modern data stack thesis was modular. Pick the best tool for each layer — Fivetran for ingestion, dbt for transformation, Looker for BI, Dagster or Airflow for orchestration — and compose them. The merger collapses the boundary between the two most common tools in the analytics engineering workflow.
This consolidation directly affects orchestration choices. dbt Cloud’s roadmap is now tied to Fivetran’s broader platform play. The combined entity has a commercial incentive to make ingestion and transformation work better together than with alternatives. That’s natural business behavior, but it means teams that rely on dbt Cloud’s built-in scheduler are now depending on a scheduling layer whose strategic priorities include selling Fivetran connectors, not just running dbt models.
Teams that want orchestration independence — the ability to choose their ingestion, transformation, and serving tools freely — need an orchestration layer they control. An external orchestrator like Dagster or Airflow sits outside the Fivetran-dbt commercial relationship. It can trigger Fivetran syncs, run dbt builds, refresh BI dashboards, and coordinate Python processing without being coupled to any single vendor’s platform strategy.
Why This Matters More Than It Sounds
Platform lock-in risk tends to feel abstract until it isn’t. Here are the concrete scenarios where orchestration independence pays for itself:
Ingestion tool switching. You’re on Fivetran today. Tomorrow, a cheaper alternative emerges that handles your specific connectors better. If your orchestration is dbt Cloud’s scheduler, switching ingestion tools means also changing how orchestration coordinates with ingestion. If your orchestration is Dagster, you swap the Fivetran sensor for an Airbyte sensor and the rest of your pipeline is untouched.
Pricing changes. The combined entity controls pricing for both ingestion and transformation. Bundled pricing can be attractive — until it isn’t. A platform-independent orchestrator ensures that a pricing change in one layer doesn’t force changes in your orchestration layer.
Feature roadmap divergence. dbt Cloud’s scheduler will evolve to serve Fivetran’s combined platform strategy. Features that benefit the combined platform (deeper Fivetran-dbt coordination) will be prioritized over features that benefit users of competing ingestion tools. If you’re using dlt, Airbyte, or custom ingestion, dbt Cloud’s scheduler investments may not align with your needs.
Consulting and advisory independence. For consultants recommending tools to clients, recommending Dagster or Airflow as the orchestration layer gives clients vendor optionality that a dbt Cloud-only approach cannot provide. You’re not tying a client to a vendor relationship that may not serve them in two years.
The Neutral Glue Layer
Both Dagster and Prefect explicitly position themselves as the neutral “glue layer” between best-of-breed tools. This positioning predates the merger, but the merger makes it more resonant.
An asset-centric orchestrator like Dagster can model your entire data platform as a unified dependency graph:
- Fivetran (or Airbyte, or dlt) syncs as upstream assets with freshness tracking
- dbt models as transformation assets with per-model lineage
- Python processing as custom assets in the same graph
- BI dashboard refreshes as downstream triggers
Every component is replaceable without restructuring the orchestration layer. The orchestrator knows about data freshness and dependencies, not about which specific vendor produced the data. That’s the architectural property that makes orchestration independence practical, not just theoretical.
Airflow achieves something similar through its provider ecosystem — 90+ provider packages covering practically every data tool. The abstraction is different (task-based rather than asset-based), but the vendor independence is the same: Airflow doesn’t care whether your ingestion is Fivetran or a custom Python script.
The Counter-Argument
To be fair: the merger could also make things simpler. A tightly integrated Fivetran-dbt platform with built-in scheduling could eliminate the need for external orchestration entirely. If the combined entity delivers on its “open data infrastructure” promise, you might get better coordination between ingestion and transformation than any external orchestrator can provide, because the platform controls both sides.
The question is whether you trust “open data infrastructure” as a durable commitment or a launch-day slogan. The history of tech consolidation suggests that platform openness tends to narrow over time as commercial pressures increase. But that’s a prediction, not a certainty.
Practical Implications
For teams evaluating orchestration in 2026 with dbt in the stack:
-
Already on dbt Cloud’s scheduler: no immediate action needed, but an exit path is worth thinking through if the platform evolves away from your needs.
-
Choosing an orchestrator for a new project: vendor independence is now a more relevant decision criterion. The orchestrator comparison covers which tool fits which team profile.
-
Consultants recommending tools: a neutral orchestration layer gives clients the ability to swap ingestion, transformation, and BI tools without cascading changes.
The merger does not make dbt Cloud’s scheduler a poor choice. It increases the strategic value of external orchestration for teams that want to preserve optionality across the stack.