How to Use DAX MAX With Filter in Power BI (12 Practical Examples)

Finding the highest value in Power BI is pretty straightforward when I only need the overall maximum. But when I want to find the highest sale for a specific region, the latest order date for a particular product, or the product associated with the most recent order, I need to apply a filter along with MAX().

In this tutorial, I’ll show you every practical way to use DAX MAX with a filter in Power BI: CALCULATE + MAX, MAXX + FILTER, ignoring slicers with ALL / REMOVEFILTERS, finding the max per group with ALLEXCEPT, and the question I get asked most often — how to return a value from another column that corresponds to the max date.

Every example uses the same small dataset, and I’ve listed the exact result each formula returns, so you can build it yourself and check your work.

Quick answer

To get the maximum value of a column for a specific condition, wrap MAX inside CALCULATE and pass the condition as a filter:

Max Sale Electronics =
CALCULATE (
    MAX ( Sales[SalesAmount] ),
    Sales[Category] = "Electronics"
)

If you need the max of a calculated expression (not a single column), use MAXX over a filtered table instead:

Max Unit Price Electronics =
MAXX (
    FILTER ( Sales, Sales[Category] = "Electronics" ),
    DIVIDE ( Sales[SalesAmount], Sales[Quantity] )
)

The rest of this post explains when to use which, and how to handle the tricky cases.

MAX vs MAXX in DAX: what’s the difference?

Both functions return the largest value, but they work differently:

	MAX	MAXX
Syntax	MAX(<column>) or MAX(<expr1>, <expr2>)	MAXX(<table>, <expression>, [<variant>])
Works on	One column (or compares two scalar values)	Any expression, evaluated row by row
Type	Aggregator	Iterator (the “X” functions)
Accepts a filter directly?	No — wrap it in CALCULATE	Yes — pass FILTER(...) as the table
Can find the max of a measure?	No	Yes, e.g. MAXX(VALUES(Sales[Region]), [Total Sales])
Blanks	Skipped in a column	Skipped
TRUE/FALSE values	Not supported (use MAXA)	Not supported

Rule of thumb: if you want the max of a single column under a condition, use CALCULATE(MAX(...), condition). If you want the max of something you calculate per row, or the max of a measure across groups, use MAXX.

Sample data used in this tutorial

Create a table called Sales in Power BI Desktop (Home > Enter data) with these 10 rows. Set OrderDate to the Date type and Quantity / SalesAmount to Whole Number.

OrderID	OrderDate	Region	Product	Category	Quantity	SalesAmount
1	2026-01-05	East	Laptop	Electronics	3	3600
2	2026-01-12	West	Phone	Electronics	5	4000
3	2026-02-03	East	Desk	Furniture	2	900
4	2026-02-18	North	Laptop	Electronics	1	1200
5	2026-03-02	West	Chair	Furniture	8	1600
6	2026-03-15	East	Phone	Electronics	4	3200
7	2026-03-28	North	Desk	Furniture	3	1350
8	2026-04-06	West	Laptop	Electronics	2	2400
9	2026-04-20	East	Chair	Furniture	6	1200
10	2026-04-20	North	Phone	Electronics	2	1600

Notice that orders 9 and 10 share the same date (2026-04-20). That’s deliberate — ties on the max date are a common source of wrong results, and I’ll show you how to handle them in Example 9.

Also create this base measure, which a few examples use:

Total Sales = SUM ( Sales[SalesAmount] )

Example 1: Basic MAX (and how filter context already affects it)

Max Sale = MAX ( Sales[SalesAmount] )

In a table visual, this returns 4000. Now put Region in a table visual next to it:

Region	Max Sale
East	3600
North	1600
West	4000
Total	4000

This is the most important thing to understand: MAX already respects filter context. Every row of the visual, every slicer and every page filter narrows down the rows MAX can see. You only need an explicit filter when you want a condition that isn’t coming from the report — or when you want to remove one.

Example 2: Power BI MAX with one filter using CALCULATE

To get the highest sale for Electronics only:

Max Sale Electronics =
CALCULATE (
    MAX ( Sales[SalesAmount] ),
    Sales[Category] = "Electronics"
)

Result in a card: 4000. By region it returns East = 3600, North = 1600, West = 4000. Change the filter to “Furniture” and the card returns 1600.

Watch out: CALCULATE overrides the existing filter on that column

Put Category in the rows of a table visual with this measure and you’ll see something surprising:

Category	Max Sale Electronics	Max Sale Electronics (KEEPFILTERS)
Electronics	4000	4000
Furniture	4000	(blank)

The Furniture row shows 4000 because the filter Sales[Category] = "Electronics" replaces the Furniture filter coming from the visual. If you want the conditions to intersect instead, wrap the condition in KEEPFILTERS:

Max Sale Electronics (KEEPFILTERS) =
CALCULATE (
    MAX ( Sales[SalesAmount] ),
    KEEPFILTERS ( Sales[Category] = "Electronics" )
)

Example 3: MAX with multiple filters in Power BI (AND / OR)

AND conditions

Separate filters in CALCULATE with commas — they’re combined with AND:

Max Sale East Electronics =
CALCULATE (
    MAX ( Sales[SalesAmount] ),
    Sales[Category] = "Electronics",
    Sales[Region] = "East"
)

Result: 3600 (orders 1 and 6 match; 3600 is the larger).

OR conditions on the same column

Use IN (or ||) when the values come from one column:

Max Sale East or North =
CALCULATE (
    MAX ( Sales[SalesAmount] ),
    Sales[Region] IN { "East", "North" }
)

Result: 3600.

OR conditions across different columns

A simple filter argument can only reference one column. For an OR across two columns, use FILTER over just those columns:

Max Sale North or Furniture =
CALCULATE (
    MAX ( Sales[SalesAmount] ),
    FILTER (
        ALL ( Sales[Region], Sales[Category] ),
        Sales[Region] = "North" || Sales[Category] = "Furniture"
    )
)

Result: 1600 (the matching rows are 3, 4, 5, 7, 9 and 10).

Example 4: MAX with a “greater than” condition in Power BI

Highest sale among orders with a quantity of 3 or more:

Max Sale Qty 3+ =
CALCULATE (
    MAX ( Sales[SalesAmount] ),
    Sales[Quantity] >= 3
)

Result: 4000. The same pattern works with >, <, <= and <> (not equal).

Example 5: Power BI MAXX with FILTER (max of a calculated expression)

Suppose you want the highest unit price sold in Electronics. There’s no unit price column, so you have to calculate it per row — that’s a job for MAXX:

Max Unit Price Electronics =
MAXX (
    FILTER ( Sales, Sales[Category] = "Electronics" ),
    DIVIDE ( Sales[SalesAmount], Sales[Quantity] )
)

Result: 1200 (the Laptop). FILTER returns only the Electronics rows, and MAXX evaluates the division on each of them and keeps the highest.

Performance tip: if you only need the max of an existing column, prefer CALCULATE(MAX(col), condition) over MAXX(FILTER(Sales, ...), col). Filtering a single column is cheaper for the engine than iterating the whole table, which matters on large fact tables.

Example 6: MAX of a measure in Power BI (highest region or product total)

MAX only accepts a column, so you can’t write MAX([Total Sales]). To find the highest total across a group, iterate the group with MAXX:

Best Region Sales =
MAXX (
    VALUES ( Sales[Region] ),
    [Total Sales]
)

Region totals are East = 8900, West = 8000 and North = 4150, so the result is 8900. Swap Sales[Region] for Sales[Product] and you get 8800 (Phone).

To return the name of the best region instead of its value:

Best Region =
VAR RegionTotals =
    ADDCOLUMNS ( VALUES ( Sales[Region] ), "@Sales", [Total Sales] )
RETURN
    MAXX ( TOPN ( 1, RegionTotals, [@Sales], DESC ), Sales[Region] )

Result: East.

Example 7: Power BI MAX ignoring filters (ALL, REMOVEFILTERS, ALLSELECTED)

Sometimes you need the overall max on every row — for example, to compare each region against the best one.

Max Sale All Regions =
CALCULATE (
    MAX ( Sales[SalesAmount] ),
    REMOVEFILTERS ( Sales[Region] )
)

% of Max Sale =
DIVIDE ( [Max Sale], [Max Sale All Regions] )

Region	Max Sale	Max Sale All Regions	% of Max Sale
East	3600	4000	90%
North	1600	4000	40%
West	4000	4000	100%

Which remover should you use?

• REMOVEFILTERS(Sales[Region]) / ALL(Sales[Region]) — ignores every filter on Region, including slicers. (REMOVEFILTERS is simply a more readable alias of ALL when used inside CALCULATE.)
• ALL(Sales) — ignores every filter on the whole table. The result is 4000 no matter what the user selects.
• ALLSELECTED(Sales[Region]) — ignores the visual’s row filter but keeps slicer selections. With a Region slicer set to East and North, it returns 3600 (the best of the selected regions), while REMOVEFILTERS(Sales[Region]) still returns 4000 from West. This is usually what users expect.

Example 8: Max per group in a calculated column (ALLEXCEPT)

If you need the max per region stored on every row — for example to flag the top order in each region — use a calculated column:

Region Max Sale =
CALCULATE (
    MAX ( Sales[SalesAmount] ),
    ALLEXCEPT ( Sales, Sales[Region] )
)

Is Top Order in Region = 
IF ( Sales[Total Sales] = Sales[Region Max Sale], "Yes", "No" )

East rows get 3600, West rows 4000, North rows 1600. Orders 1, 2 and 10 are flagged Yes. In a calculated column, CALCULATE turns the current row into a filter (context transition); ALLEXCEPT then removes every column filter except Region.

Note: Calculated columns are computed at refresh and do not react to slicers. If the “top” should change with the user’s selection, use a measure like Example 7 instead.

Example 9: Get a value from another column at the max date in Power BI

This is the most searched scenario: “return the product (or amount, or status) from the row with the latest date.” You can’t get it with MAX alone, because MAX returns a date, not the row.

Method 1: Find the max date, then filter to it

Latest Product = 
VAR LastDate1 = MAX ( Sales[OrderDate] )
RETURN
    CALCULATE (
        SELECTEDVALUE ( Sales[Product] ),
        Sales[OrderDate] = LastDate1
    )

Region	Latest Product
East	Chair
North	Phone
West	Laptop
Total	(blank)

Why is the total blank? At the total level the latest date is 2026-04-20, which has two rows (Chair and Phone). SELECTEDVALUE returns blank when there is more than one value. You have two options:

Show all tied values — replace SELECTEDVALUE(...) with:

CONCATENATEX ( VALUES ( Sales[Product] ), Sales[Product], ", " )

The total now shows Chair, Phone.

Method 2: TOPN with a tie-breaker (always returns one row)

Latest Product (TOPN) =
VAR LastRow =
    TOPN (
        1,
        Sales,
        Sales[OrderDate], DESC,
        Sales[OrderID], DESC
    )
RETURN
    MAXX ( LastRow, Sales[Product] )

The second sort key (OrderID) breaks the tie, so the total returns Phone (order 10). Region rows return the same values as Method 1. Use this method when the business rule is “the most recent record wins.”

Sum of a value on the latest date

Sales on Latest Date =
VAR LastDate = MAX ( Sales[OrderDate] )
RETURN
    CALCULATE ( [Total Sales], Sales[OrderDate] = LastDate )

Result: East = 1200, North = 1600, West = 2400, Total = 2800 (both orders on 2026-04-20).

For more date-focused patterns, see my guide on Power BI DAX Max Date.

Example 10: Max date with a condition

Latest order date for Laptops:

Last Laptop Order =
CALCULATE (
    MAX ( Sales[OrderDate] ),
    Sales[Product] = "Laptop"
)

Result: 2026-04-06. MAX works on date columns just like numbers.

Example 11: Fix the total row (sum of max values)

In Example 1, the total row showed 4000 — the max of all rows — not 3600 + 1600 + 4000. If you want the total to add up the per-region maximums:

Sum of Region Max =
SUMX (
    VALUES ( Sales[Region] ),
    [Max Sale]
)

Each region row still shows its own max, and the total now shows 9200.

Example 12: Handle blanks in MAX

MAX(column) skips blanks. If every row in the current context is blank (or there are no rows), it returns BLANK — which hides the row in many visuals. To show 0 instead:

Max Sale or Zero =
COALESCE ( [Max Sale], 0 )

Be careful with the two-argument version, MAX(expr1, expr2): it treats a blank as 0. So MAX(-5, BLANK()) returns 0, not -5 — a subtle bug when comparing negative values.

Common mistakes with DAX MAX and filters

1. Same value in every row. Your filter uses ALL(Sales) or overrides the column that’s on the visual (see Example 2). Use KEEPFILTERS or ALLSELECTED.
2. Blank total with SELECTEDVALUE. There’s a tie or multiple values at the total level — use CONCATENATEX or TOPN (Example 9).
3. MAX([Measure]) gives an error. MAX needs a column. Use MAXX(VALUES(...), [Measure]) (Example 6).
4. Calculated column doesn’t respond to slicers. That’s by design — use a measure.
5. MAX on a TRUE/FALSE column fails. Use MAXA, or convert the column to 1/0.
6. Slow reports on large tables. Replace MAXX(FILTER(FactTable, ...), col) with CALCULATE(MAX(col), column predicates) where possible.

Frequently asked questions

Conclusion

To use DAX MAX with a filter in Power BI, remember three patterns: CALCULATE(MAX(col), condition) for a single column, MAXX(FILTER(...), expression) for calculated values and measures, and “max date + CALCULATE” (or TOPN) to pull another column’s value from the latest row. Add REMOVEFILTERS, ALLSELECTED or ALLEXCEPT when you need to control which filters apply.

You may also like:

• Power BI DAX MAX Function
• Power BI DAX Max Date