Change tracking lets a client retrieve changes since a previous synchronization point rather than repeatedly exporting the full table. The client must preserve the returned continuation or delta link and correctly process deletions.
Overview
Use Dataverse change tracking and delta links to synchronize only changed rows, reduce full exports and handle deleted records in downstream systems.
Implementation guide
- Enable change tracking for the table where supported and confirm the integration identity can read the required data.
- Request the initial dataset using the supported change-tracking query pattern and store the returned delta link securely.
- On later runs, call the saved delta link exactly as returned and process changed rows and deletion markers.
- Persist the new delta link only after the batch has been processed successfully, so failures do not skip changes.
- Plan for expired synchronization tokens by detecting the documented response and performing a controlled resynchronization.
Code example
// Pseudocode for a delta synchronization loop
var url = savedDeltaLink ?? initialQueryUrl;
while (url != null)
{
var page = await dataverse.GetJsonAsync(url);
await ApplyChangesAsync(page.Value);
url = page.NextLink ?? page.DeltaLink;
}
// Persist the final delta link after successful processing.
Common pitfalls
- Do not advance the saved delta link before downstream processing succeeds.
- Do not ignore deletion markers if the target system must mirror Dataverse state.
- Plan for a full resync when a saved token is no longer valid.
Conclusion
Use the documented platform behaviour as the contract, validate the design with realistic traffic and failure scenarios, and keep operational configuration under version control where appropriate.