Einstein Discovery - Explore Story Insights

Question 1

What is an insight

Answer 1

Statistically significant finding in your data.

Provide a starting point for you to investigate the relationships among your story’s explanatory variable and its goal

Question 2

Insights Interface - Summary Panels

Answer 2

Display summation of story variables by correlation

Story Version Summary - Summary of story insights, including version comparison

Insight Summary Panels - list of variables, ordered by correlation, that positively or negatively impact a story

Question 3

Insights Interface - Variable Insights

Answer 3

Each Variable insight has the following areas:

Sensitive Field Badge - shield indicating if the variable has sensitive data

Insight title - Selected Variable’s impact on story outcome

Insight chart - Insights based on the analysis of a dataset

Version update - current version’s outcome average

Bookmark insight - save insight for future reference

Question 4

Insights Interface - If multiple insights are displayed, what order are the insights typically shown in?

Answer 4

1. Single Variable that best explains the variability of the outcome (best single predictor)
2. Variable that with the previous variable most improves the explanatory power (best two-variable predictor)
3. If appropriate, another variable that with the first variable improves the explanatory power
4. Next single variable that best explains the variability on its own (second-best single predictor)

Question 5

Insights Interface - On the Variables Panel, how are variables ordered?

Answer 5

By correlation descending

Question 6

Insights interface - Filter selector allows you to do what?

Answer 6

Surface insights by variable or value. ex - change variable and/or variable value.

Question 7

How can we interact with insights?

Answer 7

Hover over bar in graph or chart - popup will show supporting statistics for the bar.

In the insight description, click applicable hyperlink to drill down. ED redraws chart based on selection.

Question 8

What does Einstein first show you when you create or open as story?

Answer 8

Einstein shows you a list of insights, starting with the most statistically significant ones, which are the primary insights in your story. They are descriptive insights that help you explore, at an overview level, what factors contribute to the outcome.

Descriptive insights

Question 9

What do subsequent insights in the filter insight show with a selected variable?

Answer 9

Subsequent insights in the list show how combinations of values and conditions are associated with the outcome.

Question 10

What do the bars represent in a first-order insight for a numeric variable

Answer 10

Each vertical bar represents a value or bucket (range) of the Explanatory Variable.

For numeric variables with low cardinality (ten or fewer observations), Einstein Discovery displays numbers along the x-axis instead of ranges (buckets).

Question 11

What does the orange line represent in the first-order insight for a numeric variable?

Answer 11

Global average of all variables contributing to the outcome

Question 12

Bucket of the explanatory variable

Answer 12

Blue signifies a bucket that is statistically significant.

Gray signifies a bucket that is statistically insignificant.

The height of the bar shows the average Outcome Variable for this bucket relative to the Global Average.

Question 13

What are the hover-over details in the numerical chart?

Answer 13

Explanatory Variable and Bucket
Total: Sum of all Outcome Variables for this bucket.
Count: Number of observations in this bucket.
Difference from Average: How much this bucket’s average differs from the global average (Average - Global Average).
Outcome Variable: Average Outcome Variable for this bucket (Total / Count).
Global Average: Global average of all variables.

Question 14

In a first-order insight for a TEXT variable, what do the bars represent?

Answer 14

Category of that explanatory variable.

Sorted descending frequency of observations from left to right unless otherwise selected in setup.

Question 15

What does each vertical bar represent for a TEXT variable?

Answer 15

Blue signifies a category that is statistically significant.

Gray signifies a category that is statistically insignificant.

The height of the bar shows the average Outcome Variable for this category relative to the Global Average line. In this example, Division=Standard Hardware has an average CLV that is the most below the global average.

Question 16

What are the hover-over details in the TEXT Variable chart?

Answer 16

Explanatory Variable and Category
Total: Sum of all Outcome Variables for this category.
Count: Number of observations in this category.
Difference from Average: How much this category average differs from the global average (Average - Global Average).
Outcome Variable: Average Outcome Variable for this category (Total / Count).
Global Average: Global average of all variables.

Question 17

What does first-order insight for DATE Variable show?

Answer 17

Date on x-axis, and outcome variable on y-axis

Question 18

What does a second-order insight show?

Answer 18

how the combination of two explanatory variables explains variation in the outcome variable

Question 19

What is the interaction effect?

Answer 19

In second-order analysis, the combined impact of both variables together on the outcome

Question 20

What does the horizontal axis show for a second-order insight?

Answer 20

Explanatory variables for the insight, grouped. Sorted descending frequency from left to right of # obs.

Question 21

Legend element on a second-order insight

Answer 21

Describes the meaning of the vertical bars in the pairs

Category - The left bar is blue when it is statistically significant or dark gray when it is statistically insignificant.

All Other categories - the Right bar is light gray

Question 22

In the legend for a second-order insight, how does it determine whether the left bar is significant?

Answer 22

Describes the meaning of the vertical bars in the pairs.

Category - The left bar is blue when it is statistically significant or dark gray when it is statistically insignificant.
All Other Categories - The right bar is light gray.
To determine whether the left bar is significant, Einstein performs the following tests:

For the numeric use case (numeric outcome variables), Einstein performs a two-sample unpaired Student’s t-test (no assumption of equal variance) for each bar, testing it against its complement (such as Type is Consulting and Type is NOT Consulting).
For the classification use case (binary outcomes), Einstein performs a Chi-Square Goodness of Fit test to compare the observed number of the desired outcome to the expected number based on the complement.
If the p-value of the test is below 0.01, the left bar is colored blue. Otherwise, it is colored a dark gray. Blue bars, then, are statistically significantly different from their complement (that their average would be so far from their complement is unlikely to have happened randomly).

Question 23

In a second-order chart, what does the paired vertical bar represent?

Answer 23

Each pair represents a category of the explanatory variable that appears along the horizontal axis.

With each pair:

The left bar represents one category of another Explanatory Variable.

The right bar represents all other categories of this Explanatory Variable.

The height of each vertical bar shows the average Outcome Variable for this category or categories.

Question 24

Second-order chart left bar popup shows

Answer 24

First Explanatory Variable and Category

Second Explanatory Variable and Category

Difference in Average for Other Buckets: The differences between the average of records that are left bar and the right bar (left bar Average - right bar Average).

Total: Sum of all Outcome Variables for this category.

Count: Number of observations of this category.

Difference from Average: How much this category average differs from the global average (Average - Global Average).

Outcome Variable: Average Outcome Variable for this category (Total / Count).

Global Average: Global average of all variables.

Question 25

Second-order chart Right bar popup shows

Answer 25

First Explanatory Variable and Category

Second Explanatory Variable and All Other Categories

Total: Sum of all Outcome Variables for this combination of categories.

Count: Number of observations within this combination of categories.

Difference from Average: How much this combination of categories average differs from the global average (Average - Global Average).

Outcome Variable: Average Outcome Variable for this category (Total / Count).

Global Average: Global average of all variables.