I have a dataset that tracks lead generation (5 categories) to sales(binary). In between this process, there is an app involved with some analytics available - no. of sessions, duration of session, pages visited and buttons pressed (binary variable) My goal is to determine what combination of variables led to a successful sale. Here is a list of variables and type.
| # | Column | Dtype |
|---|---|---|
| 0 | id | int64 |
| 1 | sessions | int64 |
| 2 | duration | int64 |
| 3 | pages | int64 |
| 4 | Button 1 | object |
| 5 | Button 2 | object |
| 6 | Button 3 | object |
| 7 | Button 4 | object |
| 8 | Button 5 | object |
| 9 | Button 6 | object |
| 10 | Button 7 | object |
| 11 | otp | object |
| 12 | object | |
| 13 | meeting | object |
| 14 | channel | object |
| 15 | sales | object |
Will a Sankey diagram be the best way to represent this? If yes, then how? If not, then what should I try?