From data to story: a case study
1. From data to story: a case study
In this final video, we're going to put what you've learned into practice with an example case study.2. The case study: Customer Churn
The problem we're dealing with, in this case, is a recent increase in customer churn, or customers leaving a company. We're trying to find out why, so we can make recommendations to counter this increase. We've analyzed customer churn rate over time and its characteristics. Our next task is preparing a presentation for higher management.3. Formulating the central message
To start, we'll work out our central message. Why do we compose the central message first? Because it is the center of our presentation, and the other building blocks - the narrative structure and visualization - depend upon it. As before, our central message consists of three components: the problem: the increasing churn rate, our main insights: namely that the increased churn rate is related to dissatisfaction with customer service, and the impact: our specific recommendations to tackle the problem.4. Setting up the narrative structure
Now that we've drafted our central message, we're going to set up the structure of our presentation using a narrative structure. Notice how we use the parts of our central message here and expand upon it: the problem is our opening, the insights from the middle, and we end with the impact. We can show some supporting insights on our way to the central insight. For example, we could show that customer satisfaction has decreased and that most complaints are about customer service. To support these points, we make use of visualizations.5. Choosing visualizations
Now that we have prepared our central message and narrative structure, we can start thinking about which visualizations to use to support the points we want to make. We can start by visualizing our problem with the help of a line plot, showing the churn rate over time.6. Choosing visualizations
Next, we show off some of the insights we gained through our analysis. For example, one of our insights could be supported by a line plot showing customer satisfaction rate versus time.7. Choosing visualizations
A second insight could be supported by a bar plot showing count of complaints by category. These plots can be the result of our exploratory data analysis or related to the main analysis: for example; if we conducted a predictive analysis, we could show a plot of predictive power per variable to visualize the influence of each variable on the prediction results.8. Let's practice!
Now it's your turn. Have fun with your case study!Create Your Free Account
or
By continuing, you accept our Terms of Use, our Privacy Policy and that your data is stored in the USA.