Can You Export a Power BI Dashboard to Excel?

Someone asks you for "the dashboard in Excel." You open Power BI, look for an Export button that produces the thing they just described, and find that no such button exists.

That is not a permissions problem and it is not you missing a menu. Power BI genuinely has no way to hand you a dashboard as a spreadsheet, because a dashboard is not a thing a spreadsheet can hold. What it does have is three different export paths that each give you a different piece of what the person asking probably wanted — and picking the wrong one is how you end up with a 30,000-row CSV that silently dropped half the data.

This guide covers what each path actually produces, the row limits that cut exports short, and the one option most people never find that keeps the numbers connected to the source.

Can You Export a Power BI Dashboard to Excel?

Not as a dashboard. There is no command anywhere in Power BI that turns a dashboard into a workbook with the tiles, the layout, and the interactivity intact.

What you can do is export the data behind one visual at a time, and you have three ways to do it: export from a dashboard tile, export from a report visual, or connect Excel live to the underlying semantic model with Analyze in Excel.

Those three are not variations on a theme. They produce different file types, respect different row limits, and one of them stays connected to Power BI while the other two are frozen snapshots the moment they land in your Downloads folder.

Reports that look finished and are quietly incomplete are the expensive kind. Get a free audit and find out which of your numbers are actually load-bearing.

Why a Dashboard Cannot Become a Spreadsheet

A Power BI dashboard is a canvas of tiles. Each tile is a pinned snapshot of a visual that lives somewhere else — usually in a report, sometimes in several different reports built on several different semantic models.

A Dashboard Has No Single Table Underneath It

A worksheet is one rectangular grid of values. A dashboard is a collection of pointers to visuals whose underlying queries may have nothing in common with each other. There is no single table to write out, which is why the export command you are looking for was never built.

Tiles Carry Their Filters With Them

When you export from a tile, you get what that tile is showing, filters included. Microsoft's own documentation is explicit about it: "If you filtered the visualization, then the .csv export is filtered as well."

That is usually what you want. It is occasionally a trap, when the person receiving the file assumes they are looking at the full dataset and nobody told them a slicer was set.

The Layout Is the Part You Lose

If what the requester actually wanted was the *picture* — the arrangement, the colours, the KPI tiles in that order — then Excel is the wrong destination entirely, and exporting to PDF or PowerPoint will serve them better than any spreadsheet.

How to Export Data From a Dashboard Tile

This is the path most people find first, and it is also the most limited one.

Open the dashboard in the Power BI service and hover over the tile you want. In the upper right corner of the tile, open the More options (...) dropdown and select Export to .csv. The file saves to your local Downloads folder by default.

Note the format. A dashboard tile exports to `.csv` and only to `.csv`. You do not get the choice of an Excel workbook, you do not get the three data-shape options, and you are capped at the CSV row limit rather than the far larger Excel one.

There is one more wrinkle worth knowing before you promise anyone a file. If the tile came from a report carrying a sensitivity label, Power BI shows a warning, and Microsoft's advice is to go elsewhere: "To export data with a sensitivity label, open the report page that contains the original visual and export the data from there instead."

That instruction is a good default even when no label is involved. Exporting from the report gives you strictly more options than exporting from the tile.

How to Export Data From a Report Visual

Open the report rather than the dashboard, select the visual, and use the export command on the visual itself. Here you get a real choice of what shape the data arrives in.

Summarized Data

The aggregated values the visual is displaying — the numbers you can see, at the grain the visual rolled them up to. This is the right pick when someone wants to check your totals.

Data With Current Layout

The same granularity as summarized data, arranged to match the shape of the visual on screen. Microsoft notes that the granularity of these two "are identical; only the shape differs in Excel." Choose this one when the recipient wants the matrix to look like the matrix.

Underlying Data

The row-level detail behind the aggregation. This is the one people mean when they say they want "the real data," and it is also the one most likely to be unavailable to you, because it depends on your permissions to the data itself.

Most reporting problems are definition problems wearing a formatting costume. Run a free audit and get a list you can act on rather than a summary you have to interpret.

The Row Limits Nobody Reads Until It Is Too Late

This is the single most important section in this article, because exceeding these limits does not fail loudly — it hands you a file that looks complete.

For `.csv` exports, the ceiling is 30,000 rows, in both Power BI Desktop and the Power BI service. For `.xlsx` exports, it is 150,000 rows.

Matrix visuals exported with current layout are measured differently again. There the 150,000 limit applies to *data intersections* rather than rows, where intersections are rows multiplied by value columns. Microsoft's worked example: 1,000 products across 12 months with 2 measures is 24,000 intersections.

When you cross a limit, Power BI tells you — once, in a message that is easy to click past: "Exported data exceeded the allowed volume. Some data might have been omitted."

Read that sentence carefully. *Some data might have been omitted.* The file still opens. The columns still look right. The totals are simply wrong, and nothing in the workbook itself says so.

DirectQuery adds a second ceiling underneath the first: a maximum of 16 MB of uncompressed data, which can cut you off well below 150,000 rows with no round number to warn you.

Analyze in Excel: the Option Most People Miss

If the request is really "I want to work with this in Excel" rather than "I want a file," none of the above is the right answer. Analyze in Excel is.

It creates an Excel workbook containing the entire semantic model behind the report, so you can build PivotTables and Pivot Charts against the live model instead of a frozen extract. Microsoft frames the use cases as enriching report data with other assumptions, performing what-if analysis, and validating the numbers in specific visuals.

The important word is *live*. The workbook holds a connection to the semantic model and refreshes inside Excel, so it does not go stale the way an exported CSV does the instant someone adds yesterday's data.

Where to Launch It

Three places. From an open report, select Export > Analyze in Excel from the top ribbon. From the workspace, open More options (...) next to the semantic model or report and select Analyze in Excel. Or select the semantic model and use Analyze in Excel on the menu bar.

Power BI generates the workbook and saves it to your OneDrive or SharePoint account, or downloads it locally if you do not have OneDrive.

What It Requires

You need Build permission on the semantic model, or at least a Contributor role in the workspace holding it. On the licensing side you need Fabric Free, Pro, or Premium Per User — and on Fabric Free you can only work with models in My workspace or on Premium/Fabric F64 or larger capacity.

Row-level and object-level security are supported and enforced, so this is not a way around restrictions someone put on your access deliberately.

What It Will Not Do

Only Import-mode semantic models preserve hierarchies in the PivotTable field list. Direct Lake, DirectQuery and composite models do not retain theirs, which makes the field list flatter and more tedious to navigate than the report suggested it would be.

There is a size ceiling too: an error can appear if the data exceeds 2 GB, at which point the XMLA endpoint is the supported route rather than this one.

And it is not available everywhere. Power BI Report Server and Power BI Embedded with an app-owned model are out of scope, models with live connections to Azure Analysis Services or SQL Server Analysis Services are not supported, external tenant users cannot use it at all, and Mac users are limited to Excel for the web.

The report is only as trustworthy as the permissions model under it. Start with a free audit and find out where your effort is actually going.

When to Use Which

If someone wants to check a single number, export summarized data from the report visual and send the workbook.

If someone wants the detail behind a total, export underlying data — and expect it to be unavailable if you do not have rights to the raw rows.

If someone wants to keep analysing in Excel over time, use Analyze in Excel so they get a connection rather than a corpse.

If someone wants the dashboard to *look* like the dashboard, export to PDF or PowerPoint and stop trying to make Excel do it.

And if someone wants all of it, in one file, refreshing, formatted like the dashboard — tell them that file does not exist, and ask which of the four they actually need. That conversation takes two minutes and saves a day.

Limitations Worth Knowing Before You Promise a File

Export can simply be switched off. The report designer or a tenant administrator can disable it for a visual, a report, or the whole tenant, and when that happens the option is greyed out or missing entirely rather than erroring usefully.

Formatting does not survive the trip. Dynamic and visual-specific formatting — percent of grand total, for instance — is not preserved in the Excel output.

CSV exports carry no subtotal or total rows at all, which surprises people comparing an export against the visual it came from.

Dates arrive in your local machine's datetime format rather than the report designer's, which is a genuine problem when the file crosses regions.

Text columns beginning with `=`, `@`, `+`, or `-` get a leading apostrophe added, because otherwise Excel would read them as formulas.

If the model contains multiple tables with no relationships between them, only data from the first table exports.

And underlying-data export fails outright if the visual has "Show items with no data" enabled.

Long-running queries against a gateway can time out mid-export, which produces an error rather than a truncated file — annoying, but at least honest about having failed.

Sensitivity labels travel with the data. A label on the semantic model is applied automatically to the generated Excel workbook, and where the workbook already carries one, the stricter of the two wins. That is a feature rather than a bug, but it does mean an exported file can be more locked down than the person receiving it expects.

Final Thoughts

The honest answer to "can you export a Power BI dashboard to Excel" is that you can export almost everything a dashboard represents, and nothing that a dashboard *is*.

Once you accept that, the decision gets easy. Tiles give you a filtered CSV capped at 30,000 rows. Report visuals give you a choice of three shapes in a workbook capped at 150,000. Analyze in Excel gives you a live connection and the whole model. The failure mode to actually fear is not picking wrong — it is picking the CSV path for a dataset that was always too big for it, and never noticing the quiet little sentence that told you so.

If you are still mapping out which parts of the Microsoft stack live where, our guide to what the Power BI service actually is covers the service-versus-Desktop split that decides which of these export routes you even have.

Numbers that cannot be traced back to a source are decoration. Get a free audit and see which of yours hold up.