Tableau
Tableau Funtion: WINDOW_COUNT( )
Tableau Function: WINDOW_COUNT( )
Category: Table Calculation Functions
What Is the Function?
Purpose of the Function
The WINDOW_COUNT() function in Tableau is a table calculation that returns the count of non-null values within a defined window of rows in a partition. Unlike RUNNING_COUNT(), which accumulates progressively, WINDOW_COUNT() allows you to specify a custom range (window) over which the count is calculated.
In simple terms, WINDOW_COUNT() answers:
“How many non-null values exist within this specific range?”
Type of Calculations
Table calculations
Window-based aggregations
Count-based analytics
Order-dependent calculations
WINDOW_COUNT() is evaluated after aggregation and depends on partitioning, sorting, and compute direction.
Practical Use Cases
Counting records within rolling time windows
Calculating moving averages (when combined with
WINDOW_SUM())Measuring activity within recent periods (e.g., last 7 days)
Supporting rate and density calculations
Identifying window-based thresholds or milestones
WINDOW_COUNT(expression, [start, end])
| Parameter | Type | Description |
|---|---|---|
| expression | Aggregate / table calculation | The numeric or field expression to count within the window. Only non-null values are counted. |
| start (optional) | Integer | Starting row offset relative to the current row. |
| end (optional) | Integer | Ending row offset relative to the current row. |
How It Works?
Mathematical / Logical Principle
WINDOW_COUNT() calculates:
WINDOW_COUNT = COUNT(non-null values in defined window)
The window is defined using offsets relative to the current row.
Return Value
Data Type: Integer
Meaning:
Returns the count of non-null values within the defined window
Can vary by row depending on window boundaries
When Should We Use It?
Use WINDOW_COUNT() when you need to:
Count records within a sliding or fixed window
Support rolling averages or ratios
Analyze short-term trends
Build window-based statistical calculations
Normalize values within a range
Basic Usage
Count values across full partition
WINDOW_COUNT(SUM([Sales]))
Returns total non-null rows in partition
Column Usage
3-period rolling count
WINDOW_COUNT(SUM([Sales]), -2, 0)
Counts values for current row and previous two rows
Centered rolling count
WINDOW_COUNT(SUM([Sales]), -1, 1)
Counts values for previous, current, and next rows
Advanced Usage
Rolling average using window functions
WINDOW_SUM(SUM([Sales]), -2, 0)
/
WINDOW_COUNT(SUM([Sales]), -2, 0)
Calculates 3-period moving average manually
Conditional window count
WINDOW_COUNT(
IF SUM([Profit]) > 0 THEN 1 END,
-3,
0
)
Counts only rows meeting a condition
Tips and Tricks
Only non-null values are counted
Always verify Compute Using and partition settings
Use with
WINDOW_SUM()for custom averagesSensitive to sorting
Window exceeding partition boundaries is auto-adjusted
Related Functions
Functions commonly used alongside or as alternatives to WINDOW_COUNT():
RUNNING_COUNT()COUNT()WINDOW_SUM()WINDOW_AVG()SIZE()INDEX()
We’ve got plenty of resources to help you master Tableau functions. For more details, check out the official Tableau documentation. Or, if you’re ready for more practice, let’s dive into related functions and build your Tableau skills further!
If you’re ready to harness the full power of Tableau and elevate your data analytics capabilities, our expert Tableau consulting services are here to guide you. Whether you need support with building advanced calculated fields, creating dynamic visual dashboards, or optimizing your data sources for peak performance, our team of experienced Tableau consultants delivers customized solutions designed for your business needs. Visit our Tableau Consulting page to discover how we can help your organization turn data into impactful, insight-driven decisions.
1. What does WINDOW_COUNT() do in Tableau?
It counts non-null values within a defined window of rows.
2. How is WINDOW_COUNT() different from RUNNING_COUNT()?
WINDOW_COUNT() counts values within a specified range, while RUNNING_COUNT() accumulates progressively.
3. Is WINDOW_COUNT() affected by sorting?
Yes, sorting determines the order of rows in the window.
4. Does WINDOW_COUNT() include NULL values? No, only non-null values are counted.
No, only non-null values are counted.
5. Can WINDOW_COUNT() be used for rolling analysis?
Yes, by defining start and end offsets.