Skip to main content

Power BI

Files.com can collect CSV and Excel files from partners and business systems and deliver them to the storage a Power BI report already reads. Partners can keep sending files over SFTP or another supported method while the reporting team keeps its Power Query connections and refresh schedule. Files.com handles collection and delivery; Power BI interprets the files and updates its semantic model, the data behind the report.

The connection runs through the source storage. Power BI connects to the SharePoint library, Azure container, or local folder where Files.com delivers the files, using that location's address and credentials.

Source Storage and Access

Choose the destination around the report's existing source connection:

Power BI sourceFiles.com delivery
SharePoint Online document libraryConnect a SharePoint Remote Server. Select a document library or a folder inside it as the destination.
OneDrive for BusinessConnect a OneDrive Remote Server using an account with access to the drive.
Azure Blob StorageConnect an Azure Blob Storage Remote Server and deliver into the container and prefix used by the query.
Local folder or network shareUse the Files.com Agent to reach the folder. Power BI's service separately needs an on-premises data gatewayExternal LinkThis link leads to an external website and will open in a new tab that can read it.

A Sync delivers files from an intake folder to the selected storage. A Mount exposes that storage as a Files.com folder, so a partner's upload writes to the connected location. The partner's Files.com folder permissions control that upload; the Microsoft account or storage credential used by Power BI separately controls its read access. Power BI does not inherit access from a Files.com user.

With Buffered Uploads set to Automatic, Files.com receives uploads to SharePoint or OneDrive mounts before delivering them to Microsoft. A completed partner upload can precede availability at Power BI's source. Schedule the refresh after onward delivery finishes.

File Selection and Replacements

The query determines whether the input is one file replaced on each delivery or a collection of files combined over time. Microsoft's SharePoint FolderExternal LinkThis link leads to an external website and will open in a new tab and FolderExternal LinkThis link leads to an external website and will open in a new tab connectors can include files in subfolders. Keep originals, archives, and files awaiting preparation outside the selected input, or filter them out before combining. Otherwise a retained copy can become another set of rows.

Files being combined need a consistent layout. Transformations can normalize incoming data before delivery when senders use different column names or formats. Keep the delivered layout aligned with the report's query.

For a Power BI file connection using OneDrive refresh, Microsoft tracks the source item's identity. Update the existing file rather than deleting it and creating another with the same name; that replacement can break the connection. Files.com Sync replacement behavior uploads replacements to a Remote Server in place. SharePoint can also modify Office documents and cause a Sync write-back to the source, which matters when the original must remain unchanged.

Refresh Results

Power BI distinguishes OneDrive refresh from data refreshExternal LinkThis link leads to an external website and will open in a new tab. OneDrive refresh synchronizes a connected workbook or Power BI file with the service; it does not query that file's original data sources. A semantic model's data refresh reads its configured sources to update the imported data. Delivering a CSV used by a Power Query connection and replacing a connected workbook are therefore different handoffs.

Verify the whole handoff with a recognizable change in a source file. Confirm its delivery at the selected storage location, run the applicable refresh, and check Power BI's refresh history and the resulting rows or reporting period. If the file is current but the report is stale, the remaining question is which source the query reads and which refresh completed.