What This Update Actually Is
HubSpot shipped a public beta API that lets developers create and manage data sources in Data Studio entirely in code. No manual UI uploads. No CRM object requirements.
Two datasource types are supported. FILE lets you upload a CSV, XLS, XLSX, or TSV file alongside a schema definition. JSON lets you define a schema in the request body and push structured rows directly, including an incremental data-push endpoint for ongoing updates.
The full API surface covers every operation you'd need: create, retrieve, full-replace, delete, rename, and incremental row push. All of it authenticates via standard OAuth.
Why HubSpot Shipped This
Here's the problem the existing toolset created. The CRM Imports API requires data to map to CRM objects. If your data doesn't fit a contact, company, deal, or ticket, you're stuck.
We've seen this frustration across portals repeatedly. A RevOps team wants to bring in a partner enrichment feed. An ops team wants to push operational snapshots into reports. An agent-generated output needs to land somewhere it can drive a workflow. None of that maps cleanly to a CRM record.
The Data Sources API is schema-agnostic. Developers define their own columns and types. Data lands in Data Studio immediately and becomes available for datasets, reports, segments, and workflow activation. The CRM doesn't have to know about it.
The incremental data-push endpoint solves a second problem: recurring updates. Before this, teams uploading refreshed data had to replace the entire file each time. Now, for JSON datasources, you push only the new rows. Lightweight, repeatable, and far less fragile.
How to Use It Step by Step
The workflow splits based on which datasource type fits your data. Here's the practical path for each.
For a FILE datasource:
- Authenticate your app with standard OAuth and confirm access to the Data Studio scope.
- Send a multipart POST to /data-studio/data-source/2026-09-beta with your file (CSV, XLS, XLSX, or TSV) and a request body that includes the datasource name, type FILE, headerRowIndex, and a columns array with name and type for each column.
- Poll the GET endpoint and check lastIngestionStatus. Wait for SUCCESSFUL before treating the datasource as ready.
- To refresh the data later, use PUT to full-replace the datasource with a new file. Incremental push is not available for FILE type.
For a JSON datasource:
- Send a POST to /data-studio/data-source/2026-09-beta with a JSON body defining datasource name, type JSON, and a columns array. Optionally include a recordId field to enable deduplication on future pushes.
- Optionally include initial row data in the creation request, or leave the datasource schema-only and push rows separately.
- Poll GET for lastIngestionStatus: SUCCESSFUL before pushing additional rows.
- For ongoing updates, POST to /{datasourceId}/data-push with only the new rows. No file management, no full replacements required.
- Use PATCH to rename the datasource at any point. Use DELETE to remove it when it's no longer needed.
Once ingestion is confirmed, the datasource is immediately available inside Data Studio for building datasets and reports.
What It Touches in Your HubSpot Strategy
This API doesn't live in isolation. It changes how you think about data architecture inside HubSpot for any team that has been working around CRM object constraints.
Data Studio is the layer where raw ingested data becomes actionable. Once a datasource is live, it can power custom datasets, feed report charts and tables, drive list segmentation, and trigger workflow logic. That means non-CRM data can now influence how HubSpot behaves for the humans using it every day.
Key Takeaway
Schema-agnostic ingestion means your data doesn't need a CRM home first. It can arrive as a raw table and still activate segments, reports, and workflows inside HubSpot.
For RevOps teams, this opens a cleaner path for operational snapshots: inventory data, fulfillment feeds, external scoring models, or finance-side tables that previously required either a custom object workaround or a third-party data warehouse connector.
For teams building with Breeze or agentic outputs, the data-push endpoint is particularly relevant. Agent-generated data can be pushed into a JSON datasource incrementally, making it reportable and actionable inside HubSpot without human intervention on each cycle. If you're thinking about where AI outputs land in your portal, this is part of that answer.
If you're mapping out how agentic workflows and data pipelines connect inside HubSpot, the HubSpot Agentic Platform pillar gives you the strategic context to see where this API fits in the bigger picture.
On the integration side, this API changes the calculus for connected apps. Teams managing multiple data connections into HubSpot will want to audit which feeds currently rely on manual uploads or unsupported CRM object mapping. Many of those can now be replaced with a clean API pipeline.
HubSpot also recently revamped how connected apps are managed in the portal. If your team is cleaning up integrations, the Connected Apps redesign update is worth reading alongside this one.
Key Takeaway
The incremental data-push endpoint (JSON type only) is the feature that makes recurring feeds practical. Partner enrichment data, operational snapshots, and AI outputs can now update in HubSpot without replacing entire datasets each time.
Reporting is the most immediate win. Any data that lands in a Data Studio datasource is reportable right away. That means custom datasets, trend charts, and table reports can all pull from your ingested tables without requiring CRM object creation or custom property setup.
For teams who care deeply about data visibility during import cycles, pair this with HubSpot's import progress tracking update so your humans can monitor ingestion status without polling the API manually.
Who Should Care Most
This update is developer-facing. If your team doesn't have anyone writing API calls, it won't apply yet. But the downstream value touches every role on a RevOps or data team.
- HubSpot developers and integration engineers who need to move non-CRM tabular data into reports or workflows without object mapping constraints.
- RevOps and data ops leads managing partner feeds, enrichment data, or financial tables that don't belong in the CRM object model.
- Marketing ops and reporting analysts at companies that need custom dataset blending inside HubSpot's reporting layer without exporting to an external BI tool.
- Humans building AI-powered or agentic pipelines where outputs need to land in HubSpot and stay reportable and actionable.
Availability: Smart CRM Professional and Data Hub Professional. This is currently a public beta, so expect the API surface to evolve before general availability.
George's Take
What excites me about this API isn't just the technical capability. It's what it signals about where HubSpot is heading with Data Studio as a real data layer, not just a reporting skin on top of CRM records. For years, the workaround for non-CRM data was either shoehorn it into a custom object or accept that it lived outside HubSpot forever. That constraint shaped how teams built their portals and, honestly, limited what the humans running those portals could do. This API says the constraint is gone. Define your schema, push your data, use it. That's the kind of flexibility that makes HubSpot worth betting on as a true operational platform.
“For years the workaround for non-CRM data was shoehorn it into a custom object or accept it lived outside HubSpot forever. This API says that constraint is gone.”
If you're ready to map out how this API fits into your data architecture, or if you're not sure whether your current HubSpot setup is ready to take advantage of Data Studio, let's talk. A Sidekick strategy call is the fastest way to get a clear picture of what's possible in your specific portal.
Frequently Asked Questions
What is the HubSpot Data Sources API for Data Studio?
It's a public beta API that lets developers programmatically create and manage data sources in HubSpot Data Studio. It supports FILE uploads (CSV, XLS, XLSX, TSV) and JSON-based ingestion with an incremental data-push endpoint. Data does not need to map to any HubSpot CRM object type.
How is the Data Sources API different from the CRM Imports API?
The CRM Imports API requires data to map to HubSpot CRM objects like contacts or deals. The Data Sources API is schema-agnostic. You define your own columns and types, and the data lands directly in Data Studio without needing a corresponding CRM object. It also supports incremental row pushes, which the CRM Imports API does not.
What file formats does the FILE datasource support?
The FILE datasource supports CSV, XLS, XLSX, and TSV formats. You upload the file as a multipart POST alongside a schema definition. Ingestion is asynchronous, so you poll the GET endpoint for a lastIngestionStatus of SUCCESSFUL before using the datasource in reports.
What is the incremental data-push endpoint and when should I use it?
The incremental data-push endpoint is available for JSON datasources only. It lets you POST new rows to an existing datasource without replacing the entire dataset. Use it for recurring data feeds like partner enrichment data, operational snapshots, or AI-generated outputs that update on a schedule.
Who has access to the Data Sources API for Data Studio?
The API is available to accounts on Smart CRM Professional and Data Hub Professional tiers. It's currently in public beta as of July 2026, meaning all accounts at those tiers can participate but the API surface may change before general availability.
Can data ingested via the Data Sources API trigger HubSpot workflows and lists?
Yes. Once data is ingested into a Data Studio datasource and ingestion is confirmed, it becomes available for datasets, reports, list segmentation, and workflow activation inside HubSpot. Non-CRM tabular data can drive the same downstream actions as CRM-based data once it's in Data Studio.
