Live Insights

Mrunal Deshpande Updated by Mrunal Deshpande

Live Insights are fast, automated insights that help you explore and identify trend drivers, compare cohorts, and spot anomalies by automatically analyzing combinations that stand out in data.

Insights help you quickly generate powerful data analysis on the data stored in cloud data warehouses. You can run Live Insights on live connections without extracting or moving the data from its source. Live Insights are available for Trend and Comparison analysis.

With Live Insights, you can perform the following analysis on your preferred cloud data platforms:

  • Identify trend drivers: You can diagnose why important metrics and key performance indicators are changing. The system automatically scans through the underlying data and analyzes to identify the top contributors to a change and presents the findings in an understandable format.

For example: What drives change in DisplayO_TOTALPRICE between Mar 1996 and Apr 1996?

To run Live Insights for Trend analysis:

  1. Right-click on any data point in a bar or line chart created on a Live data source and select the Live tab.
  2. Under the Live tab, you will see a Trend insight created for the selected data point.
  3. Click Configure your Live Insight.
  4. From the options displayed select Trend.
  1. Provide the required fields to create Trend Analysis.

A new Live Insight will be created under the Live Insights button on the right side of the page.

  1. Once the Live Insights is ready, you can click on the See Trend Drives button to look at the top 5 contributors for the change.

  1. You can also click on View Full Insights to dig deeper into each Live Insight created. 

  • Compare cohorts: You can identify the underlying key differences between different groups and cross-sections of data by Cohort Analysis.

For example: What drives P_RETAILPRICE for Brand#35 compared to Brand#54?

To run Live Insights for Comparing Cohorts:

  1. Right-click on any data point in a bar or line chart created on a Live data source and select the Live tab.

Under the Live tab, you will see a Comparison insight created for the selected data point.

  1. Click Configure your Live Insight.
  2. From the options displayed, select Comparison.
  1. Provide the required fields to create Comparison Analysis.

A new Live Insight will be created under the Live Insights button on the right side of the page.

This Cohort Analysis compares two brands to identify the differences in prices.

  1. Once the Live Insights is ready, you can click on the See Comparison Drives button to look at the top 5 contributors for the change.

  • Anomaly Detection: Anomaly detection is a process to identify anomalies in time-series data set which differs from the norm. You can proactively uncover unexpected shifts by having the system continuously monitor data to highlight anomalies and the underlying reasons at a granular level

For example: Quantity decreased by 30.71% between Aug 1998 and Sept 1998.

To run Live Insights for Anomaly Detection:

  1. Right-click on a data point marked as anomaly in a line chart created on a Live data source and select the Live tab.
  1. Under the Live tab, you will see an Anomaly Detected insight created for the selected data point.
  2. Click Understand WHY.
  1. A new Insight for anomaly detected will be created under the Live Insights button on the right side of the page.
  1. This Trend Driver will identify the top 5 contributors for the anomaly detected in the data set.

How Did We Do?

Marketshare

Improvements to Insights Algorithms

Contact