Level 1 · What does this cell filter?

Every number in a Power BI visual is a measure evaluated under a set of filters. That set is the cell’s filter context. Reading it is the first skill of DAX: before you ask whether a formula is right, ask what the cell is filtering.

Where filters come from

A cell collects filters from everything around it:

• The visual’s own headers. A matrix row for Lighting filters products[category] to Lighting. A column header (a year across the top) adds its own filter.
• Slicers on the page. A Year slicer set to 2025 filters 'Date'[Year]. Unless you edit the interactions, a slicer filters every visual on the page.
• The Filters pane. Filters on this visual, on this page and on all pages.
• Cross-filtering. Clicking a bar in another chart filters this one too.

All of these combine with AND. A filter on category and a filter on year don’t compete; the cell keeps only rows that pass both. The Total row is the same measure with every filter except the row header, so it still respects the slicers.

How a filter reaches the numbers

Each filter lands on the table that owns the column, then travels along relationships from the one side to the many side. In the Harbor & Pine model:

• products filters order_items (one product, many order lines).
• 'Date' filters orders, and orders filters order_items.
• customers filters orders, and through it order_items.

So a slicer on 'Date'[Year] reaches order_items in two hops. The reverse is not true: with single-direction relationships, a filter on products does not reach orders. That is why “orders by category” needs DISTINCTCOUNT(order_items[order_id]) and not COUNTROWS(orders).

Where CALCULATE fits

A measure can change the context it is given. The store’s Sales measure is

Sales = CALCULATE(SUM(order_items[line_total]), orders[status] = "completed")

CALCULATE takes the cell’s filter context, adds a filter on orders[status], and evaluates the sum there. The rest of this chapter is about that: adding filters, replacing them, and removing them to compute shares and totals.

In this practice tool and in Power BI Desktop

This page runs a practice DAX engine built for this site, not Power BI. In Power BI Desktop you can see a cell’s filter context for yourself: hover the cell for the tooltip, open the Filters pane on the right to see visual, page and report filters, and use View → Performance analyzer → Copy query to see the exact filters the visual sends. On the Format ribbon, Edit interactions shows which slicers filter which visuals.

Common mistakes

• Forgetting a slicer on another part of the page. It still filters the matrix.
• Assuming filters flow both ways. A product filter doesn’t reach orders.
• Reading the Total as the sum of the rows. It is the measure evaluated again, without the row filter. For a plain SUM they agree; for averages, distinct counts and ratios they usually don’t.