Incremental copy gets more flexible: New watermark column types in Copy job in Fabric Data Factory (Generally Available)

[yexu]

If you haven’t already, check out Arun Ulag’s hero blog “FabCon and SQLCon 2026: Unifying databases and Fabric on a single, complete platform” for a complete look at all of our FabCon and SQLCon announcements across both Fabric and our database offerings. 

---

Introduction

Copy job is the go-to solution in Microsoft Fabric Data Factory for simplified data movement across multiple clouds. With native support for bulk copy, incremental copy, and change data capture (CDC) replication, it can handle a wide range of movement scenarios through an intuitive, easy-to-use experience.

Incremental copy is a foundational pattern for keeping analytics systems up to date—but in practice, not all source systems expose changes in the same way. Many real-world databases don’t use a single, clean datetime or INT column to track updates.

Today, we’re introducing expanded watermark column type support in Copy job, making incremental copy more flexible and applicable to a wider range of source systems.

What’s new: More watermark column types

Copy job now supports three more watermark column types for incremental copy:

• ROWVERSION
• Date
• String (interpreted as datetime)

This enhancement allows you to choose the column that best represents change in your source system—while Copy job continues to manage state, checkpoints, and incremental windows automatically.

Why these new watermark column types matter

ROWVERSION: Precise change tracking in SQL systems

Many SQL-based systems expose a ROWVERSION column that automatically changes whenever a row is modified.

With ROWVERSION as a watermark:

• Every insert or update is captured reliably.
• There’s no dependency on application-managed timestamps.
• It’s ideal for high-throughput transactional systems.

Copy job uses the monotonic nature of ROWVERSION values to track progress safely across runs.

Date: Simple and explicit time-based incremental copy

Date or datetime columns remain one of the most common patterns for incremental ingestion.

With native Date watermark support:

• Use existing LastUpdatedDate, ModifiedAt, or similar columns.
• Copy job automatically applies delayed extraction to ensure there is no data loss or overlap between runs, safely managing incremental windows.

String (interpreted as datetime): When datetime aren’t stored as datetime

In some systems, timestamps are stored as strings rather than true datetime types.

Copy job now supports string-based watermark columns if the value can be interpreted as a datetime.

This means:

• There’s no need to cast or transform columns into custom queries.
• No schema changes are required in the source.
• Incremental copy works even when timestamps are stored as strings.

It’s a small change with a big impact on real-world compatibility.

Getting started

There’s nothing new to enable. When configuring incremental copy in Copy job:

• Select your source table.
• Choose an incremental column.
• Pick ROWVERSION, Date, or String (datetime) based on your schema.

Copy job handles the rest—tracking progress, managing windows, and ensuring exactly-once semantics.

Learn more

What is Copy job?

Submit feedback on Fabric Ideas and join the conversation in the Fabric Community.

Questions or feedback? Leave your thoughts in the comment section.