CHAPTER 1 ■ STATISTICS AND PROBABILITY
2
The cycle sharing scheme provides means for the people of the city to commute
using a convenient, cheap, and green transportation alternative. The service has 500
bikes at 50 stations across Seattle. Each of the stations has a dock locking system (where
all bikes are parked); kiosks (so customers can get a membership key or pay for a trip);
and a helmet rental service. A person can choose between purchasing a membership
key or short-term pass. A membership key entitles an annual membership, and the key
can be obtained from a kiosk. Advantages for members include quick retrieval of bikes
and unlimited 45-minute rentals. Short-term passes offer access to bikes for a 24-hour
or 3-day time interval. Riders can avail and return the bikes at any of the 50 stations
citywide.
Jason started this service in May 2014 and since then had been focusing on
increasing the number of bikes as well as docking stations in order to increase
convenience and accessibility for his customers. Despite this expansion, customer
retention remained an issue. As Jason recalled, “We had planned to put in the investment
for a year to lay out the infrastructure necessary for the customers to start using it. We
had a strategy to make sure that the retention levels remain high to make this model self-
sustainable. However, it worked otherwise (i.e., the customer base didn’t catch up with
the rate of the infrastructure expansion).”
A private service would have had three alternatives to curb this problem: get
sponsors on board, increase service charges, or expand the pool of customers. Price hikes
were not an option for Jason as this was a publicly sponsored initiative with the goal of
providing affordable transportation to all. As for increasing the customer base, they had
to decide upon a marketing channel that guarantees broad reach on low cost incurred.
Nancy, a marketer who had worked in the corporate sector for ten years, and Eric, a
data analyst, were explicitly hired to find a way to make things work around this problem.
The advantage on their side was that they were provided with the dataset of transaction
history and thus they didn’t had to go through the hassle of conducting marketing
research to gather data.
Nancy realized that attracting recurring customers on a minimal budget
required understanding the customers in the first place (i.e., persona). As she stated,
“Understanding the persona of your brand is essential, as it helps you reach a targeted
audience which is likely to convert at a higher probability. Moreover, this also helps in
reaching out to sponsors who target a similar persona. This two-fold approach can make
our bottom line positive.”
As Nancy and Eric contemplated the problem at hand, they had questions like the
following: Which attribute correlates the best with trip duration and number of trips?
Which age generation adapts the most to our service?
Following is the data dictionary of the Trips dataset that was provided to Nancy and
Eric: