Driver Lifetime Value

Lyft: Segment drivers by lifetime value using K-Means clustering and retention analysis. Python, Pandas, Scikit-learn, Seaborn.

Assignment

After exploring and analyzing the data, please:

1. Recommend a Driver's Lifetime Value (i.e., the value of a driver to Lyft over the entire projected lifetime of a driver).
2. Please answer the following questions:
   • What are the main factors that affect a driver's lifetime value?
   • What is the average projected lifetime of a driver? That is, once a driver is onboarded, how long do they typically continue driving with Lyft?
   • Do all drivers act alike? Are there specific segments of drivers that generate more value for Lyft than the average driver?
   • What actionable recommendations are there for the business?
3. Prepare and submit a writeup of your findings for consumption by a cross-functional audience.

You can make the following assumptions about the Lyft rate card:

• Base Fare $2.00
• Cost per Mile $1.15
• Cost per Minute $0.22
• Service Fee $1.75
• Minimum Fare $5.00
• Maximum Fare $400.00

Data Description

You'll find three CSV files attached with the following data:

driver_ids.csv

• driver_id Unique identifier for a driver
• driver_onboard_date Date on which driver was on-boarded

ride_ids.csv

• driver_id Unique identifier for a driver
• ride_id Unique identifier for a ride that was completed by the driver
• ride_distance Ride distance in meters
• ride_duration Ride duration in seconds
• ride_prime_time Prime Time applied on the ride

ride_timestamps.csv

• ride_id Unique identifier for a ride
• event describes the type of event; this variable takes the following values:
  • requested_at - passenger requested a ride
  • accepted_at - driver accepted a passenger request
  • arrived_at - driver arrived at pickup point
  • picked_up_at - driver picked up the passenger
  • dropped_off_at - driver dropped off a passenger at destination
• timestamp Time of event

You can assume that:

• All rides in the data set occurred in San Francisco
• All timestamps in the data set are in UTC

Practicalities

Please work on the questions in the displayed order. Make sure that the solution reflects your entire thought process - it is more important how the code is structured rather than the final answers. You are expected to spend no more than 2 hours solving this project.