ClearInsights Max: Change Analysis

Change Analysis allows organizations to compare any two data points in a visualization for change and identify key change drivers from the underlying attribute columns. In this article, we dive into change analysis and how it is used with ClearInsights.

In this article:
Change Analysis Overview
How to run a Change Analysis
How Change Analysis calculates outliers

Change Analysis Overview

Change Analysis compares data points of simple or complex measures using these aggregates:

1. Sum
2. Count
3. Sum over sum, which generates 'what if' percentage insights
4. Average, which generates 'what if' percentage insights
5. Other functions that use a 'versus' analysis to show the absolute change grouped by different drill attributes.

How to run a Change Analysis

Contextual change analysis

ClearInsights supports contextual change analysis for area, line, column, bar, line-column, donut, and KPI charts. To perform this analysis, follow these steps:

1. Hold shift and click on two points.
2. Right-click and choose “Run change analysis.”
   
   • Note: If you are analyzing a KPI chart, you must either click the percent change label or select two points from the sparkline visualization. You cannot run change analysis on a KPI chart with no sparkline visualization.
3. A modal appears with the results of your analysis. By default, ClearInsights picks the top 5 most relevant columns, based on what it learned from your past activity.
4. To select other columns to analyze, click Customize Attributes in the lower left corner of the window and select the desired column from the window that appears.
5. You can now examine each visualization of the analysis for details of comparison.

---

Keep in Mind

1. You can add the specific visualization to an existing Liveboard or create a new Liveboard and add it there.
2. You can run another round of analysis, download the data as a CSV file, edit this visualization, or make a copy of the visualization.
3. When you run a change analysis, ClearInsights detects additional insights and displays them below the change analysis visualizations. These insights display values that underwent the highest percentage change between the data points defined in your analysis, and values whose share of the total changed the most.
   • Note: Insights do not appear for measures of type unique counts and ratios.

Iterative Change Analysis

You can now drill down on the results of change analysis to learn more about change drivers in your data.

To run an iterative change analysis, follow these steps:

1. Select two points on a chart and right-click to select Run change analysis.
2. Right-click on an attribute value in the resulting change analysis and select Analyze.
3. ClearInsights adds the specified attribute value as a filter and re-calculates change analysis for the original comparison points.

Iterative change analysis does not work on visualizations created with custom calendars.

How Change Analysis Calculates Outliers

Z-Score

A Z-score represents the number of standard deviations from the mean a data point is. Z-scores are a way to compare results from a test to a “normal” population. The basic Z-score formula for a sample is: z = (x – μ) / σ where

• σ is the standard deviation
• μ is the mean
• x is the data value

Currently, any value outside μ +/- N * σ (i.e. Z-score less than or greater than N) is an anomaly. N is determined based on the number of rows. It is chosen such that the max or min has < 0.1 probability of crossing the threshold. Change Analysis uses the following table to approximately choose the value of N, that is, how many standard deviations we want to be able to use as threshold:

Cardinality	N (number of standard deviations from mean)
>= 10000	2
< 20	3
< 500	4
< 10000	5