Power BI filters restrict which rows a visual, page or report shows, applying at three levels: visual, page and report, each narrowing what viewers can see. Use the Filter Pane for structural or background rules, and slicers for the two or three inputs viewers need to change most often.

What Are Power BI Filters?

A filter hides rows that do not meet a condition. Unlike a calculated column, which changes the data itself, a filter only changes what is visible for a given view.

Filters can be applied in three primary ways:

1. The Filter Pane. A side panel where the report designer or a viewer applies constraints.

2. Slicers. On-canvas visuals that let a viewer filter data interactively.

3. Cross-highlighting. Clicking a data point on one visual to filter the others on the page.

The Three Levels of Power BI Filters (Scope)

Visual Level Filters

These apply to a single visual only.

• Use case: show the top 5 products in one bar chart without affecting the rest of the page.

• How to use it: select the visual, open the Filter Pane, and drag a field into Filters on this visual.

Page Level Filters

These apply to every visual on the current page.

• Use case: a page dedicated to a single year's sales. A Year filter on the page makes every visual on it respect that constraint automatically.

• How to use it: drag a field into Filters on this page.

Report Level Filters

These apply to every page in the report.

• Use case: a report built for one region. A report-level filter makes every page show only that region's data, regardless of which page the viewer opens.

• How to use it: drag a field into Filters on all pages.

Report-level filters are easy for a viewer to forget exist. If someone can't find data that should be there, check for one before assuming the data itself is missing.

Slicers vs. the Filter Pane

Slicers sit on the report canvas and are the more interactive, more visible option, but they take up screen space. The Filter Pane is collapsible and supports more advanced options, such as Top N and advanced text or number conditions, but is less visible to a casual viewer. Use slicers for the two or three inputs viewers change most often, such as date range or region, and the Filter Pane for everything else. Our filters vs. slicers guide covers the full comparison and when to use each.

Advanced Filtering Types

Top N Filtering

Filters for the top or bottom items by a measure, such as the top 10 customers by total revenue. Top N is available only as a visual-level filter.

Drillthrough Filters

Lets a viewer click a data point to navigate to a detailed target page. Power BI carries the filter context from the source page to the target page automatically.

Relative Date Filtering

A built-in filter type for phrases like last 30 days, this calendar year, or next 2 weeks. It suits an operational dashboard that needs to stay current without a manual filter change.

DAX-Driven Filtering

The Filter Pane and slicers cover most cases, but some filtering logic needs DAX. The CALCULATE function is the core tool: it overrides or adds to the filter context a measure evaluates in.

Sales LY = CALCULATE(
    [Total Sales],
    SAMEPERIODLASTYEAR('Date'[Date])
)

CROSSFILTER and KEEPFILTERS let you change how relationships filter each other, or protect an existing filter from being overwritten by a new one.

You can also build a filter measure that returns 1 or 0, then apply it as a visual-level filter. For example, flagging customers whose sales grew more than 10% against the prior year:

High Growth Flag =
VAR CurrentSales = [Total Sales]
VAR PriorSales = [Sales LY]
VAR GrowthRate = DIVIDE(CurrentSales - PriorSales, PriorSales)
RETURN
    IF(GrowthRate > 0.1, 1, 0)

Cascading Filters (Dependent Slicers)

A cascading filter is when the selection in one slicer limits the options shown in another, such as picking Germany in a Country slicer and having a City slicer narrow to Berlin, Munich and Hamburg. This works naturally when relationships run from the one-side to the many-side in a clean star schema.

Best Practices for Using Power BI Filters

• Do not overload the canvas with slicers. Ten or more slows rendering and confuses users; move secondary filters into the Filter Pane instead.

• Lock and hide structural filters, such as one removing null rows, so a user cannot accidentally remove them.

• Use a hierarchy slicer when filtering by category and subcategory, to save canvas space.

• Watch filtering on high-cardinality columns. Filtering on a column with millions of unique values can slow a report down.

• Keep bidirectional cross-filtering as the exception, not the default; it can cause ambiguous results and slower performance. See our cross-filter direction guide for when bidirectional filtering is actually the right call.

• For a report that needs to change which measure or dimension a visual shows, not just which rows are visible, use a field parameter instead of a filter. See our field parameters guide for how to build one.

Filters are additive: a report-level filter, a page-level filter, a slicer and a DAX measure can all apply to the same visual at once. If a number looks wrong, check each level starting from the visual and working up.

FAQs

Can I turn off cross-filtering between visuals?

Yes. Use Edit Interactions on the Format ribbon to stop one visual from filtering another.

What is the difference between a filter and a slicer in Power BI?

A slicer is an on-canvas visual a viewer interacts with directly. A filter is a background rule applied through the Filter Pane. Slicers take up canvas space; filters do not.

How do I clear all filters in Power BI?

Add a "Clear all slicers" bookmark button to the report, so a viewer can reset their view with one click.

Sources

• Filter types in Power BI reports - Microsoft Learn

• Drillthrough in Power BI Desktop - Microsoft Learn