Rolling Window Tool
The Rolling Window tool is a powerful feature that brings advanced row-level calculations to the Cloud Native experience. It enables dynamic, sliding-window style aggregations—ideal for time-series, performance monitoring, and any analysis requiring context over sequential rows of data.
Whether you're analyzing trends, monitoring real-time performance, or building time-based metrics, this tool provides the flexibility to compute context-aware, rolling aggregations over your dataset.
Notice
The Rolling Window tool is available only as a part of Live query.
The Rolling Window Tool unlocks rich, contextual insights by enabling you to compute metrics over moving windows. Common use cases include:
Financial and Time-Series Analysis: Compute rolling averages, sums, or ranks over the last N rows or time periods. Also, identify short-term volatility in stock prices or trading volumes.
Sales and Business Insights: Detect trends with 5-day, weekly, or monthly moving averages, and track performance of KPIs within custom periods.
Performance and Operations Monitoring: Monitor team or system performance over the last 7 days and capture real-time operational trends using dynamic windows.
Configuration
To begin, you have two column creation options. You can either create a new column to store the results of your calculation or update an existing column by overwriting its values.
You need to specify a name for the new output column that will contain the rolling calculation results.
After naming the column, select an aggregate function. Available options include average, sum, minimum, maximum, count, rank, and others. The function you choose will determine how the values within the defined window are summarized.
Select the numeric column on which the aggregation will be performed.
Following that, set the window range to define how the rolling window operates:
From The First Row
Until The Last Row
N Rows Before Current Row
N Rows After Current Row
Current Row
From First Row To N Rows After Current Row
N Rows Before Current Row Until Last Row
N Rows Before And After Current Row
You can control the order of data before the window is applied by sorting it based on one or more columns.
For example, sorting by "Hobbyist" and then by a time or ID column ensures that rows are sequenced properly.
Optionally, you may define columns to group by. This applies the rolling window within specific subgroups of your dataset, which is ideal for performing partitioned aggregations, such as calculations per country or per respondent.
Tip
Sorting is required to apply the rolling logic correctly. Ensure your data is sorted in a meaningful order (e.g., chronologically, by user ID, etc.) to get accurate results.