Salesforce DevOps for Data Cloud

Most Data Cloud configuration lives outside the Metadata API. Here's what that means for your release process.

What's hard to deploy

Data Streams, Data Model Objects, Calculated Insights, and Segments mostly live in Data Cloud's own configuration layer rather than the standard Metadata API, so a change set or 2GP package rarely moves them cleanly between orgs. Sandbox refreshes don't carry Data Cloud's ingested data with them either, which means teams often rebuild or reconnect data streams by hand in every new sandbox instead of promoting a tracked change.

Where it gets hard

Core objects live outside the Metadata API

Data Streams, Data Model Objects, Calculated Insights, and Segments mostly live in Data Cloud's own configuration layer, so a change set or 2GP package rarely moves them cleanly between orgs.

Sandbox refreshes don't carry ingested data

A sandbox refresh resets the org but not Data Cloud's ingested records, so teams often rebuild or reconnect data streams by hand in every new sandbox instead of promoting a tracked change.

Ingestion connectors need reconnecting per org

Data Cloud's source connectors are configured and authenticated per environment, so even a clean org copy still needs someone to reconnect every source before data starts flowing.

In practice

How Serpent helps

Serpent centralizes every sandbox and scratch org your Data Cloud setup depends on, syncing environments from one place so the Salesforce-side metadata around it, permission sets, flows, integration users, stays version-tracked even though Data Cloud's own configuration lives outside the Metadata API. See org management in Serpent for how environments stay in sync. For teams evaluating a dedicated data-movement tool, see how Serpent compares to Prodly.

Typical release for Data Cloud

Task the Salesforce-side change
Permission sets, flows, and the integration user config a Data Cloud stream depends on go into one task.
Delta deploy the supporting metadata
Serpent ships just what changed, not a full org deploy, so the Data Cloud objects you can't move stay isolated from what you can.
Sync environments before reconnecting streams
Pooled sandboxes and scratch orgs stay aligned from one dashboard, so reconnecting a data stream in a new environment starts from a known-good baseline.
Roll back the metadata, not the pipeline
If something breaks, a one-click rollback reverts the Salesforce-side change without touching Data Cloud's own configuration.

Common questions

Data Cloud DevOps, answered

Does Serpent deploy Data Streams, DMOs, or Segments directly?

No tool moves Data Cloud's native configuration through the Metadata API, that layer lives outside it entirely. Serpent keeps the surrounding Salesforce metadata and environments synced so the rest of your setup doesn't drift.

Why does a sandbox refresh break our Data Cloud setup?

A refresh resets the org but not Data Cloud's ingested data or stream connections. Serpent's org pooling gives you a consistent baseline to reconnect from instead of starting over each time.

Can Serpent track the integration users and permission sets Data Cloud depends on?

Yes. Those are standard Salesforce metadata, so they deploy and version through Serpent's normal pipeline alongside everything else.

Start free. No credit card, no install, no commitment.

Set up in under 15 minutes. No DevOps hire needed.

Start free Book a demo