Sign In
Get Started

Using the FILTER Function in DAX – A Detailed Guide With Examples

by | Power BI

Table of Contents

Introduction to the FILTER Function in DAX

What is the FILTER Function in DAX?

The FILTER function in Data Analysis Expressions (DAX) is a powerful and versatile tool used to return a table that contains a subset of the original table based on a given expression or condition. It is widely used in Power BI, SQL Server Analysis Services (SSAS), and other tools that support DAX.

Syntax

FILTER(<table>, <expression>)
  • <table>: The table to be filtered.
  • <expression>: A logical expression that defines the conditions for rows to be included in the returned table.

Example 1: Basic Usage

Assume we have a table named Sales with the following columns: ProductID, Quantity, SalesAmount.

Sales Now On Advertisement

To filter the Sales table to include only rows where Quantity is greater than 10, you can write:

FILTER(Sales, Sales[Quantity] > 10)

Example 2: Using FILTER with CALCULATE

CALCULATE is often used in combination with FILTER to adjust the context of a calculation. For instance, to calculate the total SalesAmount for transactions where Quantity is greater than 10, use:

TotalFilteredSalesAmount := CALCULATE(
    SUM(Sales[SalesAmount]),
    FILTER(Sales, Sales[Quantity] > 10)
)

Example 3: Combining FILTER with Multiple Conditions

To filter the Sales table for rows where Quantity is greater than 10 and SalesAmount is greater than 500, you can write:

FILTER(Sales, Sales[Quantity] > 10 && Sales[SalesAmount] > 500)

Example 4: Filtering Related Tables

Assume we have another table named Products with columns: ProductID, Category, ProductName.

To filter the Sales table to include only rows where the related Products[Category] is “Electronics”, use:

FILTER(
    Sales,
    RELATED(Products[Category]) = "Electronics"
)

Example 5: Using FILTER with ALL to Remove Context Filters

To calculate the total SalesAmount ignoring existing filters on ProductID, you can use:

TotalSalesIgnoringProductFilter := CALCULATE(
    SUM(Sales[SalesAmount]),
    FILTER(ALL(Sales[ProductID]), Sales[Quantity] > 10)
)

Conclusion

The FILTER function in DAX is essential for creating customized subsets of data relevant to specific conditions or requirements. It becomes significantly more powerful when combined with other DAX functions like CALCULATE or RELATED. By mastering the use of FILTER, practitioners can create more dynamic and insightful data models.

Power BI Tools Advertisement

A Comprehensive Guide to Understanding and Implementing the FILTER Function in DAX

Simple Examples and Basic Use Cases

1. Filtering a Single Column

Example: Filtering a Sales Table for Sales Greater than $500

FILTER(Sales, Sales[Amount] > 500)

This DAX expression filters rows in the Sales table where the Amount column is greater than 500.

2. Filtering with Multiple Criteria

Example: Filter Sales Greater than $500 and Region is ‘North’

FILTER(Sales, Sales[Amount] > 500 && Sales[Region] = "North")

This filters the Sales table to include rows where Amount is greater than 500 and Region is ‘North’.

3. Combining FILTER with CALCULATE

Example: Total Sales Amount for Specific Criteria

CALCULATE(
    SUM(Sales[Amount]),
    FILTER(Sales, Sales[Amount] > 500)
)

This expression calculates the total amount of sales where the Amount is greater than 500.

4. Using FILTER Inside Other Functions

Example: Average Sales Amount for a Specific Product

CALCULATE(
    AVERAGE(Sales[Amount]),
    FILTER(Sales, Sales[ProductID] = "P123")
)

This calculates the average sales amount for the product with ProductID ‘P123’.

5. Filtering on Dates

Example: Sales in the Year 2022

FILTER(Sales, YEAR(Sales[OrderDate]) = 2022)

This filters sales records to include only those where the order date falls in the year 2022.

6. Combining FILTER with ALL

Example: Removing All Filters on a Table Before Applying New Filters

CALCULATE(
    SUM(Sales[Amount]),
    FILTER(ALL(Sales), Sales[Amount] > 1000)
)

This removes any existing filters on the Sales table before applying a new filter for sales amounts greater than 1000.

7. Filtering Related Tables

Example: Filtering Sales Based on a Product Attribute

CALCULATE(
    SUM(Sales[Amount]),
    FILTER(
        Sales,
        RELATED(Product[Category]) = "Electronics"
    )
)

This calculates the total sales amount for products in the ‘Electronics’ category, using the RELATED function to access the Category column in the Product table.

Advanced Filtering Techniques in DAX

In this section, we’ll explore advanced filtering techniques using the FILTER function in DAX. We will cover complex scenarios where FILTER is combined with other functions to achieve sophisticated filtering.

Using FILTER with CALCULATE

The CALCULATE function is often used in combination with FILTER to modify the context in which a calculation occurs. Below is an example demonstrating how to use these functions together:

Total_Sales_Filtered =
CALCULATE(
    SUM(Sales[Amount]),
    FILTER(
        Sales,
        Sales[Year] = 2023 && Sales[Region] = "North America"
    )
)

This expression calculates the total sales amount for the year 2023 in the North America region.

Using FILTER with RELATEDTABLE

The RELATEDTABLE function returns a table that is related to the current table, enabling you to filter across relationships.

High_Revenue_Customers =
CALCULATE(
    COUNTROWS(Customers),
    FILTER(
        RELATEDTABLE(Sales),
        Sales[Amount] > 10000
    )
)

This calculates the number of customers who have made individual purchases exceeding $10,000.

Using FILTER with ALL

The ALL function removes all filters from the specified columns or the entire table. This is useful for creating calculations that need an unfiltered context.

High_Sales_No_Filter =
CALCULATE(
    SUM(Sales[Amount]),
    FILTER(
        ALL(Sales),
        Sales[Amount] > 5000
    )
)
'''

This calculates the total sales amount for transactions greater than $5000 without considering any existing filters on the `Sales` table.

## Using FILTER with VALUES

The `VALUES` function returns a single-column table of unique values. Combining it with `FILTER` enables dynamic filtering based on unique entries.

```dax
Customer_Sales_Rank =
CALCULATE(
    RANKX(
        ALL(Customers),
        CALCULATE(SUM(Sales[Amount])),
        ,
        DESC
    ),
    FILTER(
        VALUES(Customers[CustomerID]),
        Customers[Region] = "Europe"
    )
)

This rank calculation determines the sales rank of each customer within the European region.

Using FILTER with EARLIER

The EARLIER function returns the value of an outer row context. It is particularly useful in row context manipulations like nested row calculations.

Data Mentor Advertisement 
Cumulative_Sales =
SUMX(
    FILTER(
        Sales,
        Sales[Date] <= EARLIER(Sales[Date])
    ),
    Sales[Amount]
)

This calculates the cumulative sales amount up to the current date for each row in the Sales table.

Implementing Dynamic Date Filtering

Dynamic date filtering can be achieved through the DATESBETWEEN and TODAY functions to maintain filters relevant to the current date.

YTD_Sales =
CALCULATE(
    SUM(Sales[Amount]),
    DATESBETWEEN(
        Sales[Date],
        DATE(YEAR(TODAY()), 1, 1),
        TODAY()
    )
)

This expression calculates the year-to-date sales amount.

Combining Multiple Filters

Apply multiple conditions by nesting or chaining FILTER functions.

Complex_Filter_Sales =
CALCULATE(
    SUM(Sales[Amount]),
    FILTER(
        FILTER(
            Sales,
            Sales[SalesRep] = "John Doe"
        ),
        Sales[OrderDate] > DATE(2023, 1, 1) && Sales[OrderDate] < DATE(2023, 12, 31)
    )
)

This expression computes the sales amount for John Doe within the year 2023.

By mastering these advanced filtering techniques in DAX, you can perform complex data analysis and gain deeper insights from your data models.

Combining FILTER with Other DAX Functions

In this section, we will demonstrate how you can combine the FILTER function with other DAX functions for more advanced data analysis. This guide intends to show practical examples where the power of FILTER combined with other functions is fully illustrated.

Example 1: CALCULATE and FILTER

Scenario

You want to calculate the total sales for products with sales over 1000 units.

Implementation

TotalSalesOver1000Units = 
CALCULATE (
    SUM ( Sales[SalesAmount] ),
    FILTER ( Sales, Sales[UnitsSold] > 1000 )
)

Explanation

  • CALCULATE: Evaluates an expression in a modified filter context.
  • SUM ( Sales[SalesAmount] ): Sum of sales amounts.
  • FILTER ( Sales, Sales[UnitsSold] > 1000 ): Filters the Sales table to include only records where UnitsSold is greater than 1000.

Example 2: FILTER and ALL

Scenario

You need to calculate the percentage of total sales that come from a specific category, disregarding the current filter context.

Implementation

CategorySalesPctOfTotal = 
DIVIDE (
    SUM ( Sales[SalesAmount] ),
    CALCULATE (
        SUM ( Sales[SalesAmount] ),
        ALL ( Sales )
    )
)

Explanation

  • DIVIDE ( , ): Safely divides the first argument by the second.
  • SUM ( Sales[SalesAmount] ): Sales amount for the current filter context.
  • CALCULATE ( SUM ( Sales[SalesAmount] ), ALL ( Sales ) ): Sum of all sales, ignoring any filters currently applied to the Sales table.

Example 3: FILTER and RELATED

Scenario

You want to get the total sales amount for products from a specified product category.

Implementation

TotalSalesByCategory = 
CALCULATE (
    SUM ( Sales[SalesAmount] ),
    FILTER (
        Sales,
        RELATED ( Products[Category] ) = "Electronics"
    )
)

Explanation

  • RELATED ( Products[Category] ): Retrieves the category from the related Products table.
  • FILTER ( Sales, RELATED ( Products[Category] ) = "Electronics" ): Filters the Sales table to include only rows where the related product category is “Electronics”.

Example 4: FILTER and EARLIER

Scenario

You need to calculate the rank of each product based on units sold within each category.

Implementation

ProductRankInCategory = 
RANKX (
    FILTER (
        Products,
        Products[Category] = EARLIER ( Products[Category] )
    ),
    Products[UnitsSold],
    ,
    DESC
)

Explanation

  • RANKX ( , , , DESC ): Ranks products based on UnitsSold in descending order.
  • FILTER ( Products, Products[Category] = EARLIER ( Products[Category] ) ): Filters the Products table within each category context.
  • EARLIER ( Products[Category] ): References the current row’s category value.

By combining the FILTER function with other DAX functions, you can achieve complex and powerful data analytics capabilities within your models. These advanced DAX queries allow you to address various practical business requirements efficiently.

Real-World Applications and Best Practices of the FILTER Function in DAX

Practical Implementation

Sales Analysis by Department

Objective

Analyze total sales for different departments within a retail store, filtering out only the departments with sales greater than $100,000.

DAX Formula

TotalSalesByDepartment = 
CALCULATE(
    SUM(Sales[Amount]),
    FILTER(
        Sales,
        Sales[Department] IN {"Electronics", "Clothing", "Home Goods"} && Sales[Amount] > 100000
    )
)

Customer Segmentation Based on Purchase Frequency

Objective

Segment customers based on the number of purchases they have made, categorizing them into ‘Low’, ‘Medium’, and ‘High’ frequency segments.

DAX Formula

CustomerSegment = 
CALCULATE(
    COUNT(Sales[CustomerID]),
    FILTER(
        Sales,
        Sales[CustomerID] = Customers[CustomerID] &&
        RELATED(Customers[PurchaseFrequency]) IN {"Low", "Medium", "High"}
    )
)

Impact of Marketing Campaigns on Sales

Objective

Determine the sales attributed to a particular marketing campaign.

DAX Formula

CampaignSales = 
CALCULATE(
    SUM(Sales[Amount]),
    FILTER(
        Sales,
        Sales[CampaignID] = Campaigns[CampaignID] &&
        Campaigns[CampaignName] = "Spring Sale"
    )
)

Best Practices

Use CALCULATE for Context Transition

Always wrap your FILTER function within CALCULATE for it to be able to alter the context, ensuring accurate results based on specific conditions.

Minimize Row Context

Avoid overly complex row contexts within FILTER that involve multiple lookups or relationships, as they can degrade performance. Simplify where possible.

Use Variables for Readability

When dealing with long or complex FILTER conditions, use variables to hold partial results. This enhances readability and maintenance.

VAR MinSales = 100000
RETURN
CALCULATE(
    SUM(Sales[Amount]),
    FILTER(
        Sales,
        Sales[Amount] > MinSales
    )
)

Test and Validate Filters

Regularly test FILTER conditions to ensure they are correctly applied, especially when combining multiple conditions or using them in advanced calculations.

Example Combination with Other Functions

Combining FILTER with SUMX

Objective

Calculate the weighted average price of products within a category.

EDNA AI Advertisement

DAX Formula

WeightedAvgPrice = 
CALCULATE(
    SUMX(
        Products,
        Products[Price] * Products[SalesQuantity]
    ) / SUMX(
        Products,
        Products[SalesQuantity]
    ),
    FILTER(
        Products,
        Products[Category] = "Electronics"
    )
)

Combining FILTER with DISTINCT

Objective

Count distinct customers who made a purchase within the last year.

DAX Formula

DistinctCustomersLastYear = 
CALCULATE(
    DISTINCTCOUNT(Sales[CustomerID]),
    FILTER(
        Sales,
        Sales[PurchaseDate] >= TODAY() - 365
    )
)

Combining FILTER with RELATEDTABLE

Objective

Summarize the total number of orders by customers from a specific city.

DAX Formula

TotalOrdersFromCity = 
CALCULATE(
    COUNTROWS(RELATEDTABLE(Orders)),
    FILTER(
        Customers,
        Customers[City] = "New York"
    )
)

With these examples and best practices, you can effectively implement and leverage the FILTER function in DAX for real-world applications, ensuring both performance and accuracy in your data models.

Related Posts

Comprehensive Data Analysis using Power BI and DAX

Data Model Discovery Library

An interactive web-based application to explore and understand various data model examples across multiple industries and business functions.