Hi Power Platform team & community — I’m running into a serious limitation when my data source is Excel files on SharePoint Online (or Google Sheets) and I need to refresh large datasets in Power BI / Paginated Reports.
Problem summary
The existing refresh/connectors force downloading the full file or only allow small-scale refreshes, which makes it impractical for millions of rows.
The “refresh dataset” / dataset REST API approach effectively handles only small batches (I hit practical limits ~1k rows when trying to simulate incremental behavior).
Paginated Reports let you pass parameters, but they only work reliably for fields that are present in the report visuals. If you have multiple slicers/filters (say 5 filters) and two of those fields are NOT included in the visual, you cannot parameterize those fields easily.
Paginated Reports also do not support hierarchical filter structures well (e.g., Country → State → City) for parameterized refresh scenarios.
The net result: no easy, low-code connector to do server-side / source-side filtering, pagination and partial/incremental refresh against file-based sources like SharePoint Excel or Google Sheets.
What I’m asking for (feature request)
Please build a first-party connector or improve existing connectors for Excel (SharePoint/OneDrive) and Google Sheets that provides:
Server-side filtering & query folding for file-based sheets — allow Power BI/flows to push filters to the source so only matching rows are transferred.
Incremental refresh support for sheet sources — enable RangeStart/RangeEnd style incremental behavior without downloading whole file.
Parameterized refresh independent of visuals — allow paginated reports / dataset refresh APIs to accept parameters for any column (not just those in the visual) and apply them server-side.
Hierarchical parameter support — built-in support for hierarchical filters (parent→child) in the parameter UI so paginated reports and dataset refresh can accept and apply them.
Efficient pagination & batching — built-in pagination with controllable page size (e.g., 5k/10k), retry/backoff handling, and resumable batches.
Connector UX — simple configuration to map sheet → table, define key column for incremental logic, and enable scheduled or ad-hoc parameterized refresh via API.
Support for Google Sheets with equivalent capabilities (not just Excel/SharePoint).
Why this matters
Many customers keep source data in Excel/Google Sheets (non-DB) by necessity. Moving everything to a SQL-like store is not always feasible. A robust connector would enable:
real incremental refresh without architecture overhaul,
lower egress costs and faster refreshes,
simpler paginated report automation and parameter-driven refreshes, and
better support for reporting teams that rely on spreadsheets.
Ask to product team / community
Is there any roadmap for adding server-side filtering / incremental capabilities to file-based connectors?
If a first-party connector is not planned, are there recommended workarounds (patterns, sample flows, or Azure Functions templates) that scale to millions of rows without full-file downloads?
Upvote / comment if you also need this capability — it would help prioritize a proper connector.
Thanks — happy to provide sample scenarios, JSON of my current flow, or test cases if useful.
Share
Report
All responses (
Answers (
52,857
