Make Data-Driven Decisions in Customer Success (2nd Ed.)

This video introduces the CSMath learning series. It begins by introducing the Law of Large Numbers and describes how measuring churn is like flipping a coin. The video introduces the concepts of the Confidence Level and the Confidence Interval and shows how typical logo churn measurements are full of random errors. The exercises show how to install the Excel Data Analysis ToolPak and provides practice interpreting Confidence Intervals.

As a result of this module, students will:

• Understand that a very large number of samples are required for resolving small differences in means

• Learn most senior leaders make important decisions based on chance, rather than facts, and that these decisions impact people's lives

• Appreciate that in order to become truly "data driven," it's essential to understand natural variation and account for it

• See the value of completing the CSMath course   

This lecture describes data types, sets, samples and populations. The exercises provide practice defining sets using common Excel features.   

As a result of this module, students will: 

• Understand the differences between a set, a population, and a sample

• Be able to sort, filter, and use Pivot Tables to define sets

• Describe different data types and which pertain to Customer Success

• Identify methods to prepare sets for analysis

This lecture describes statistics and frequency distributions, paying particular attention to the Standard Normal.

As a result of this module, students will:

• Be able to compute mean, median, mode, range and standard deviation for variable and categorical data using Excel functions

• Describe the difference between parameters and statistics

• Be able to construct a histogram

• Use the Normal Deviate to estimate probabilities

• Be able to determine if a distribution is normally distributed or skewed

NOTE: Be sure to download the CSMath Handy Equations and Procedures Guide for this lesson and for those that follow.

This lesson covers hypothesis testing and statistical errors.

As a result of this lesson, students will be able to:

• Use a 5-step method to determine if chance can be ruled out when comparing means

• Determine if an upper-tailed, lower-tailed, or two-tailed test is appropriate

• Use the correct statistical tests given data types and distribution assumptions

• Describe the differences between Type I and Type II errors, including which is generally more concerning

This lesson describes more sensitive and accurate methods to track logo churn, revenue churn, expansion revenue, and other key metrics.

As a result of this lesson, student will be able to:

• Describe fundamental problems with how financial people measure churn

• Describe the benefits of control charts

• Select the appropriate control chart type

• Use the procedure to set up and interpret control charts

This lesson describes tools to study how factors relate to outcomes using correlation, contingency tables, and ANOVA.

As a result of the lesson, the student will be able to:

• Use Excel statistical functions and the Data Analysis ToolPak to calculate correlation, chi-square, and one-way ANOVA

• Describe the difference between correlation and causation

• Apply hypothesis testing to rule out chance associations

• Construct contrasts (test group vs. control group) and use ANOVA to analyze results

This lesson gives an overview of regression analysis and mathematical modeling.

As a result of the lesson, students will be able to:

• Use simple and multiple linear regression capabilities in Excel to screen factors and construct predictive models

• Determine the quality of fit between the mathematical model and the data

• Construct more predictive customer health dashboards

• Identify situations when logistic regression or the Generalized Linear Model are better solutions

This module covers short-term and long-term forecasting methods.

As a result of this lesson, students will be able to:

• Construct sales forecasts from regression models

• Make a case for less tampering with the numbers

• Fit an exponential model

• Apply an Excel tool for classical forecasting

In this lesson, we describe how to use data-driven decision making within the context of continuous improvement.

As a result of this lesson, students will:

• Challenge conventional wisdom using outcomes metrics to evaluate individual contributor performance

• Examine the fundamental factors that affect all Customer Success processes

• Be able to apply a very common process improvement framework

• Beware the human proclivity to favor intuition over logic

Note: There are no exercises with this lesson. Proceed to the quiz after watching the video.

In this lesson, we'll discuss Bayes' Theorem and give an overview of Predictive Analytics.

As a result of this final lesson in the series, students will be able to:

• Update probabilistic forecasts with new information

• Share the benefits of Predictive Analytics

• Share the pitfalls that come with Predictive Analytics