Netflix Case Study: Analyzing Customer Behavior and Market Trends

This case study examines Netflix's evolution in the video rental industry, starting with its DVD-by-mail service and its transition to video streaming. It explores the impact of technological advancements on the industry, including the rise of kiosks and video-on-demand services. The case analyzes Netflix...Read More

1 Netflix: The Customer Strikes Back
Introduction
Three years after earning his MBA, Hunter Keay was starting to make a name for
himself at a leading investment bank when, in February 2012, some of his clients
grew increasingly anxious about the value of their holdings in Netflix, Inc. (Netflix),
the subscription-based media distribution company. Six months earlier, Netflix had
announced a plan to split its on-demand video streaming and DVD mail delivery into
two businesses and to increase the price of its most popular service. But in the face of
near-universal criticism, Netflix had abandoned the plan within a month, only to lose
800,000 subscribers and half its stock value (Figure 11-1). Keay’s clients who held
Netflix wanted to know what remained of their investment.
Figure 11-1 Netflix stock price and volume, March
2002 to February 2012
Source: Yahoo! Finance.
To determine a more accurate value of Netflix stock, rather than apply one of the
standard methods favored by his firm, Keay was considering the use of customer
lifetime value (CLV). He was not certain that the metric applied in this instance,
whether the firm even considered it valid, or how CLV related to the more accepted
methods. He was certain about one thing, though: New technologies were
transforming the industry and the ways customers received video content. The
question was whether “Netflix 2: The Sequel” would ever be as popular as the
original.
An Industry Driven by Technology
The video rental industry has been substantially altered by technological
developments outside the industry. Major milestones included the DVD by mail that
could be ordered via the Internet, video streaming, and lately kiosks.

The Traditional Retail Rental Store
The advent of videotape, acceptance of the VHS cassette standard, and subsequent
affordability of home videocassette players in the 1980s brought with them the
proliferation of the movie rental business. By the 1990s, the majority of market
share had consolidated to a few participants with similar business models
competing on selection, price, and especially location. National chains, such as
Blockbuster and Hollywood Video, grew by staking claims at strategic locations
with adequate population density. By 1990, Blockbuster professed to have a store
within a ten-minute drive of 70% of the U.S. population. Mom-and-pop video stores
survived by finding locations the chains did not seek.
Movie rental required that a customer leave his or her home with the intention of
renting, then make a spontaneous decision based on what was available. The cost of
a video rental ranged from $3.00 per week for older movies to $6.00 per three days
for new releases (allowing for weekend viewing when rented on Friday, the most
popular day). Small mom-and-pop stores typically had a collection of a few
hundred videos for rental; a Blockbuster store had about 2,500 titles. A store’s
video paid for itself after 13 rentals, so films with mass appeal were the norm;
nearly 70% of all films rented at Blockbuster were new releases. Limited selection
and stock-outs were a common concern, as was the relative convenience of store
hours.
Late returns were a thorny problem: A movie could not be rented until it was back
on the shelf, and a scarcity of titles might deter a customer from returning. So video
stores charged late fees, which monetized the delay and encouraged the customer to
return movies promptly. In reality, as one commentator noted, late fees called
attention to customer failure, in the manner of “a disapproving librarian tallying up
35 cents in overdue fines while floating the unspoken accusation you were
irresponsible on top of everything else.”1 When Blockbuster eventually dropped
many forms of late fees, the move resulted in a charge to revenue of $400 million.
The bricks-and-mortar value proposition was eroding.
DVD by Mail
DVD mail service started to gain popularity in the early 2000s. The subscribing
customer selected a movie on a website, and a DVD would arrive at his or her home
in about one business day. The customer could keep the DVD as long as he or she
liked, then mail it back to the provider in the envelope provided. By selecting
multiple movies and arranging them in order of priority in an online queue, the
customer could ensure prompt delivery of subsequent selections and always have
something on hand to watch as opportunities arose. Subscription tiers were based on
how many movies a customer could receive simultaneously and priced accordingly,
starting at $7.99 per month for one movie at a time. (See Exhibit 11-1 for a co
(Venkatesan 144-146)

Venkatesan, Rajkumar, Paul Farris, Ronald Wilcox. Cutting Edge Marketing Analysis:
Real World Cases and Data Sets for Hands on Learning. Pearson Learning
Solutions, 06/2014. VitalBook file.
DVD by Mail
DVD mail service started to gain popularity in the early 2000s. The subscribing
customer selected a movie on a website, and a DVD would arrive at his or her home
in about one business day. The customer could keep the DVD as long as he or she
liked, then mail it back to the provider in the envelope provided. By selecting
multiple movies and arranging them in order of priority in an online queue, the
customer could ensure prompt delivery of subsequent selections and always have
something on hand to watch as opportunities arose. Subscription tiers were based on
how many movies a customer could receive simultaneously and priced accordingly,
starting at $7.99 per month for one movie at a time. (See Exhibit 11-1 for a
complete pricing comparison.)
Video on Demand
Video on demand (VOD) was content distribution via an Internet-connected
television, computer, or mobile device. The customer selected a movie from an
online menu and, within seconds, the movie began streaming to his or her device.
The customer could view the content as it was downloaded, rather than waiting for
the complete file, which otherwise could take almost as long as the running time of
the film. No exchange of a data-storage medium was required, so stock-outs and
late fees were avoided, and a significantly larger and more eclectic catalog could be
offered.
Kiosk Rentals
Movie rental kiosks were freestanding dispensers of DVDs located in high-traffic
areas with extended—sometimes 24-hour—access, such as convenience stores,
grocery stores, and fast-food restaurants. Redbox, the dominant player, founded in
2003, was originally funded by McDonald’s. As of 2012, Redbox claimed to have
rented 1.5 billion movies from 30,000 kiosks nationwide and to operate a kiosk
within a five-minute drive of two thirds of the U.S. population. Its only significant
competitor, albeit a much smaller player, was Blockbuster’s “Blockbuster Express”
kiosks.
Kiosks revolutionized the rental price point (about $1.00 per night per movie) and
changed consumer renting behavior by eliminating the planning ahead required by
DVD-by-mail services and the need to go to another location required by rental
stores. Plus, 24-hour access freed customers from time constraints. Selection,
however, was limited by two major shortfalls: the physical space inside the kiosk

and delayed releases to kiosks by movie studios wary of cannibalizing DVD sales.
A New Range of Business Models
As content delivery methods increased, an industry participant could employ different
pricing heuristics across different channels and different end-user content licenses. As
such, revenue model, delivery method, and content licensing were dimensions by
which each participant might be assessed (see Table 11-1).
In terms of revenue, a business was either pay-per-view or monthly subscription.
Depending on the delivery method, the one-time fee of the pay-per-view model
would entitle the customer to rent one DVD by mail or online streaming access for a
finite time period. In the case of purchase, a one-time fee entitled the buyer to
indefinite ownership of streaming content or of an actual DVD.
Table 11-1 Perceptual Market Map for the VHS and
Digital Eras
(Venkatesan 146-147)
Venkatesan, Rajkumar, Paul Farris, Ronald Wilcox. Cutting Edge Marketing Analysis:
Real World Cases and Data Sets for Hands on Learning. Pearson Learning
Solutions, 06/2014. VitalBook file.
Content was either delivered by physical DVD or streamed over the Internet from the
service’s website to the user’s computer or ancillary television device, sometimes
called a streaming player. Physical discs were still the dominant medium, but
increased digital access was expected to continue (Figure 11-2). The downward
pressure on physical discs was somewhat mitigated by the increasing popularity of
kiosk rental systems such as Redbox. A user’s right to content varied by service
provider and plan, but generally fell into one of three categories: rental for a finite
time period, outright purchase for unlimited personal use, or access to an entire online
library from which content could be streamed.
Figure 11-2 Digital streaming as percentage of
content delivery, 2005 to 2015 (projected)
(Venkatesan 148)
Venkatesan, Rajkumar, Paul Farris, Ronald Wilcox. Cutting Edge Marketing Analysis:
Real World Cases and Data Sets for Hands on Learning. Pearson Learning
Solutions, 06/2014. VitalBook file.

Source: Mintel/Digital Entertainment Group, May 2011.
Netflix: “Delivering Goosebumps”
Reed Hastings founded Netflix in 1997 in Los Gatos, California, after paying $40 in
late fees to the local video store for Apollo 13, and later asking, “How come movie
rentals don’t work like a health club, where, whether you use it a lot or a little, you
get the same charge?”2 The key was to let people watch movies whenever they
wanted. The Netflix model was simple: Movies that consumers ordered from
Netflix’s website were shipped to their houses. Once consumers watched the movies,
they returned them to Netflix in envelopes that were shipped along with the DVDs.
Netflix claimed that it could ship videos to most customers in less than 24 hours.
Netflix’s first innovation, in December 1999, was to eliminate late fees. Customers
paid a fixed monthly fee of about $16, rented as many as four movies in a single
order, and kept films as long as they wanted. Technically, the longer customers kept
films, the lower Netflix’s shipping cost per rental. Customer retention under this
system, however, depended on customers renting more movies per month: the more
rentals per month, the more value customers placed on the service. As Hastings
stated. “If they [the customers] rent just two movies a month, they may decide it is
not worth it.”3 This made Netflix’s movie recommendation system extremely
important: Good recommendations increased queue length, which increased retention,
which increased customer lifetime value.
To expand its customer base and reduce its reliance on the most popular films, Netflix
invested significantly in data mining technology. Netflix developed a simple but
effective movie recommendation algorithm that compared each user’s purchase to
those of customers with similar tastes and then suggested films that were highly rated
and unseen. These reviews, together with a catalog of close to 85,000 titles, held new
releases to only 30% of rentals, and 95% of Netflix’s titles were rented every quarter.
Netflix was picking up revenue from a far broader distribution of preferences than a
retail store could ever offer.
As the Netflix catalog grew, the recommendation system became simpler and more
robust. In January 2000, Netflix introduced a new simple and accurate

recommendation system called CineMatch. Each customer was prompted to rate
certain movie genres and specific movies on a one- to five-star scale. The program
found others in its database with similar preferences and then offered a predicted star
value for each movie. As the customer rated more films, the accuracy of the data
improved substantially. As Hastings stated, “Over 50% of our traffic comes via the
recommendation system. It requires a lot of database work done in real time.”4 By
2007, Netflix had close to one billion movie reviews, with customers reviewing an
average of 200 movies each.
CLV depended on the extent to which Netflix could leverage its large catalog by
encouraging customers to rent more. Its target for per-customer monthly orders was
five, the corporate average. Special promotions encouraged current customers to refer
the service to friends and family; efforts resulted in an upward trend in customer
retention (Figure 11-3).
Figure 11-3 Paid subscribers and retention rate,
March 2001 to December 2011
Source: Netflix Q1earnings report, 2012.
Conclusion
Keay asked his analyst to compile the financial data required to calculate CLV at
Netflix, but a considerable amount of work lay ahead. Was CLV an appropriate
approximation of firm value in this setting? Does CLV track with market
capitalization, discounted cash flows, or other traditional firm valuation techniques?
How sensitive was CLV to various operational and strategic changes? And what role
might be played by changes in technology? With such a public call on a volatile stock
at this point in his young career, Keay could not afford to miss.
(Venkatesan 148-150)
Venkatesan, Rajkumar, Paul Farris, Ronald Wilcox. Cutting Edge Marketing Analysis:
Real World Cases and Data Sets for Hands on Learning. Pearson Learning
Solutions, 06/2014. VitalBook file.