Data Tool Guides

Tools

9 Best Dataverse ETL Tools for 2026

Compare 9 Dataverse ETL tools, including Azure Synapse Link and Fabric Link, for moving Power Apps and Dynamics 365 data into any warehouse.

October 9, 2026

Last updated October 13, 2026

Summarize with AI:

TL;DR

Short answer: Microsoft has retired two native Dataverse export routes in a row, Data Export Service and then Export to Data Lake, and now points to Azure Synapse Link and Fabric Link. Those suit you if your destination is Azure or Fabric. If it is Snowflake, BigQuery or anything else, or you want control over the tables you end up with, Dataverse ETL tools read the Web API directly instead. Nine options:

Airbyte: open-source Dataverse source with incremental sync, landing data in any of 700+ connectors' destinations, inside Azure or out.
Fivetran: reaches Dataverse through its Dynamics 365 CRM connector, managed and billed on monthly active rows.
Azure Synapse Link and Fabric Link: Microsoft's native export to Azure Data Lake or Microsoft Fabric. No connector to run, Azure only.
Azure Data Factory: Microsoft's managed pipeline service with a native Dataverse connector.
Matillion: a Dynamics 365 query component, with transformation pushed into your warehouse.
Microsoft SSIS: reaches Dataverse through a third-party adapter, included with SQL Server licensing. Windows-bound.
Informatica PowerCenter: enterprise ETL with Dynamics connectivity, though 10.5 left standard support in March 2026.
Pentaho Data Integration: visual ETL reaching Dataverse through REST and OData steps.
Apache Airflow: an orchestrator. It schedules the Dataverse Web API calls you write yourself.
One Dataverse constraint: service protection limits cap each user at 6,000 Web API requests per five-minute window, shared with every app and flow running as that user. Run extraction under a dedicated application user, and expect a large initial load to be throttled rather than fast.

Dataverse ETL tools move data out of Microsoft Dataverse, the storage behind Power Apps and Dynamics 365, into a warehouse where it can be analysed alongside everything else. The hard parts are specific to Dataverse: an API that pages and throttles, choice columns that come back as integer codes, and lookups that need resolving before the data makes sense.

Why is exporting from Dataverse hard?

Microsoft has changed the native export route twice. Data Export Service, which replicated Dataverse into Azure SQL, was deprecated in November 2021 and reached end of life a year later. Its replacement, Export to Data Lake, was itself deprecated on 1 November 2024. Microsoft now points customers to Azure Synapse Link for Dataverse and Fabric Link, which work well inside Azure but assume your destination is Microsoft's and give you limited control over the resulting tables.

The data itself has traps too. Choice columns are stored as integer codes, so a tool that does not fetch the labels leaves you with numbers. Lookups store the related record's ID, and polymorphic lookups such as Customer or Regarding can point at more than one table. Incremental sync relies on change tracking, which has to be enabled per table.

How do the Dataverse ETL tools compare?

Tool	How it reads Dataverse	Destinations	Hosting	Pricing basis
Airbyte	Pre-built source over the Web API, incremental with change tracking	Any	Cloud or self-hosted	Free self-hosted; Cloud volume-based; Pro and Flex capacity-based
Fivetran	Dynamics 365 CRM connector, all Dataverse objects	Any	Cloud	Monthly active rows
Synapse Link and Fabric Link	Native continuous export	Azure Data Lake or Microsoft Fabric	Microsoft-managed	Azure or Fabric consumption
Azure Data Factory	Native Dataverse connector	Strongest in Azure	Azure	Per activity and data movement
Matillion	Dynamics 365 query component	Snowflake, Databricks, Redshift	Your cloud account	Credits
Microsoft SSIS	Third-party Dataverse adapter	Mostly SQL Server	On-premises Windows or Azure	SQL Server licence plus adapter
Informatica PowerCenter	Dynamics connectivity	Broad	On-premises	Quoted licence
Pentaho Data Integration	REST and OData steps you configure	Broad	Self-hosted	Community and enterprise editions
Apache Airflow	Your own Web API code	Whatever you write	Self-hosted or managed	Free, you run it

Which Dataverse ETL tools should you consider?

Two of these are Microsoft's own routes and work best inside Azure; the rest read the Dataverse Web API and can land data anywhere.

1. Airbyte

Airbyte is an open-source data integration platform with 700+ connectors and a community of 25,000+ data engineers. Its Microsoft Dataverse source authenticates through a Microsoft Entra ID app registration, reads tables over the Web API, and syncs incrementally on tables with change tracking enabled. It loads into any Airbyte destination, so Dataverse can land in Snowflake or BigQuery as easily as in Azure.

What's unique about Airbyte?

Connectors are open source and editable, and the Connector Development Kit and no-code Connector Builder cover anything without a pre-built connector. Dataverse usually sits beside other SaaS and database sources, and Airbyte moves them through the same pipeline. It runs as self-hosted open source, on Airbyte Cloud, or as Enterprise Flex.

2. Fivetran

Fivetran is a closed-source, managed ELT service with 700+ connectors. It has no separately named Dataverse source: its Microsoft Dynamics 365 CRM connector reads through Dataverse and covers standard and custom tables. It keeps choice columns as integer codes, so you join the labels yourself, and it bills on monthly active rows.

Here is a closer look at Airbyte vs Fivetran.

3. Azure Synapse Link and Fabric Link

These are Microsoft's current native routes. Azure Synapse Link for Dataverse continuously exports selected tables to Azure Data Lake Storage Gen2, optionally converting them to Delta Lake and connecting a Synapse workspace. Fabric Link makes Dataverse tables available in Microsoft Fabric without you managing the copy. Both are configured from the Power Apps maker portal with no connector to run. The catch is that your destination has to be Azure or Fabric, and you pay for the storage and compute they consume.

4. Azure Data Factory

Azure Data Factory is Microsoft's managed pipeline service, with a native Dataverse connector for copy activities. It suits teams who want Dataverse moving on a schedule alongside other Azure pipelines and need more control over transformation than Synapse Link gives. Pricing is per activity and data movement, and it is strongest when the destination is also in Azure.

5. Matillion

Matillion reads Dataverse through its Dynamics 365 query component and then transforms inside Snowflake, Databricks or Redshift, with dbt support since v1.70. It bills in credits tied to job runtime and makes most sense if Matillion already runs your warehouse transformations.

Here is a closer look at Airbyte vs Matillion.

6. Microsoft SQL Server Integration Services (SSIS)

SSIS ships with SQL Server but has no built-in Dataverse source, so teams add a third-party adapter for Dynamics 365 and Dataverse. It is a common pattern in organisations that already replicate Dynamics into SQL Server, and costs nothing beyond SQL Server and the adapter licence. It is Windows-bound and weak on cloud destinations, and most new work goes to Azure Data Factory instead.

7. Informatica PowerCenter

Informatica PowerCenter is long-established enterprise ETL with connectivity to Microsoft Dynamics and Dataverse, common in estates that already run Informatica. Two things to weigh: Salesforce completed its acquisition of Informatica in November 2025, and PowerCenter 10.5 left standard support in March 2026, so new investment is going into Informatica's cloud platform.

8. Pentaho Data Integration

Pentaho Data Integration, long known as Kettle, can call the Dataverse Web API through its REST and OData steps. You handle authentication, paging, throttling and choice labels yourself, so it suits teams already running Pentaho rather than ones choosing a tool for Dataverse.

9. Apache Airflow

Apache Airflow is an open-source orchestrator, not an ETL tool. You can schedule Python code that calls the Dataverse Web API, but you write the authentication, paging, throttling and incremental logic yourself. Its real role here is scheduling a connector-based tool such as Airbyte alongside your other pipelines.

Here is a closer look at Airbyte vs Airflow.

Should you use Azure Synapse Link for Dataverse or an ETL tool?

Use Azure Synapse Link or Fabric Link when your analytics already live in Azure or Fabric. They are Microsoft's supported route, they export continuously, and there is no connector to maintain.

Use an ETL tool that reads the Dataverse Web API when your warehouse is outside Azure, such as Snowflake or BigQuery, when you want Dataverse in the same pipeline as your other sources, or when you want to decide how choice labels, lookups and deleted records are represented rather than inheriting Microsoft's export format. Some teams combine the two: Synapse Link into a lake, then a general tool from there.

Whichever you choose, the Dataverse Web API exposes standard and custom tables, choice and lookup metadata, and change tracking for incremental reads. Authentication is through a Microsoft Entra ID app registration mapped to a Dataverse application user, and that user's security role decides which tables and rows come back.

How do you start syncing Dataverse data with Airbyte?

Step 1: Set up Dataverse as a source

Register an app in Microsoft Entra ID and create a client secret, then add it in the Power Platform admin center as an application user with a security role that can read the tables you need. Enable change tracking on those tables for incremental sync. In Airbyte, open Sources, choose New source and select Microsoft Dataverse, then enter your environment URL, tenant ID, client ID and client secret, and run the connection test.

Step 2: Set up a destination

Choose where the data lands: a cloud warehouse such as Snowflake, BigQuery or Azure Synapse, a data lake, a database or cloud storage.

Step 3: Configure the connection

Select the tables to replicate, use incremental sync where change tracking is enabled, and schedule the first full load outside business hours: even a dedicated application user is throttled at 6,000 requests per five minutes, so a large initial load takes time.

The process is the same in Airbyte Open Source, which you can deploy yourself, and in Airbyte Cloud, which you can try free for 14 days.

Which Dataverse ETL tool should you choose?

If your analytics live in Azure or Fabric, start with Synapse Link or Fabric Link and add Azure Data Factory where you need scheduled transformation. If your warehouse is anywhere else, use a tool that reads the Web API: Airbyte if you want Dataverse alongside your other sources, self-hosted or in Cloud, or Fivetran through its Dynamics 365 CRM connector if you prefer it managed. Keep SSIS or Informatica where they already run, and run any of them under a dedicated application user so extraction does not compete with your apps for API capacity.

For tools that cover more than Dataverse, see our guide to the top data integration tools.

Sync Dataverse anywhere with Airbyte

Move Dataverse and Dynamics 365 tables into Snowflake, BigQuery, Azure or any Airbyte destination with incremental sync, self-hosted or in Airbyte Cloud.

Try Airbyte Cloud free for 14 days or talk to our team.

Frequently Asked Questions (FAQs)

1. What replaced Export to Data Lake for Dataverse?

Azure Synapse Link for Dataverse, which exports tables to Azure Data Lake Storage Gen2, and Fabric Link, which makes them available in Microsoft Fabric. Export to Data Lake was deprecated on 1 November 2024, following Data Export Service, which reached end of life in November 2022.

2. What are the Dataverse API limits?

Service protection limits allow each user 6,000 Web API requests per five-minute sliding window, alongside limits on combined execution time and concurrent requests. Exceeding them returns throttling errors, so ETL tools should run under a dedicated application user and back off when throttled.

3. Can you sync Dataverse to Snowflake or BigQuery?

Yes, with an ETL tool that reads the Dataverse Web API, such as Airbyte or Fivetran. Microsoft's native Synapse Link and Fabric Link only export to Azure Data Lake or Microsoft Fabric.

4. Why do Dataverse choice columns show up as numbers?

Dataverse stores choice values as integer codes, with the labels held separately in metadata. Unless your tool fetches the labels, you need to load the choice metadata and join it in the warehouse to get readable values.

9 Best Dataverse ETL Tools for 2026

Related posts