Video Rental Developments and the Supply Chain - John M. Olin

Document Sample
Video Rental Developments and the Supply Chain - John M. Olin Powered By Docstoc
					Video Rental Developments and the Supply Chain:
Netflix, Inc*.

In an age where everything from dinner to dry cleaning to prescription medication can be
delivered to your home, it should come as no surprise that the entertainment field is
following suit. While in-home on-demand movie entertainment is certainly not new—
VCRs have been around since 1976 and by 2000 nearly 90% of homes with televisions
also owned VCRs5 —the advent of DVD technology along with sophisticated internet
commerce has allowed for changes in the way rentals, and rental businesses, work.

Traditional Rental Stores

        Traditional video rental, perhaps best illustrated by the ubiquitous video store
Blockbuster, Inc. (BBI), involves brick and mortar stores located in strategic locations,
each staffed by about a dozen employees, carrying about 1000 titles in both VHS and
DVD format4. Members arrive at the location, make selections from the available stock,
pay for their selection, about $4 for a new release, $2 for older movies and children’s
movies, and return home to watch their selection. After the designated rental period the
member returns his selection or incurs late fees. These late fees account for 18-20% of
revenues in traditional video rental stores5.

        Traditional video stores base their inventory decisions on formulas using
historical rental data, store size, and box office sales among other factors. As titles
become less popular, inventory can be reduced by selling previously viewed copies to
members for a low cost. Titles that are out of stock are often compensated with coupons
good for a free rental when it is available again. Revenue sharing models, pioneered by
Blockbuster and now common throughout the industry, lowered the cost of VHS
inventory allowing for higher rental volume at a lower risk to the storeowner, but also set
minimum and maximum levels for inventory and demand a percentage of rental revenue
for the first six months of release. DVDs are traditionally sold to rental agencies at the
same price consumers see, a low “sell-through” price, allowing for lower inventory costs
without the revenue sharing contracts. Blockbuster owns all of its DVDs outright.
However, even with higher inventories, rentals are limited by physical inventory and
excess inventory can still be sold at a loss.

  This case was prepared by Julie Niederhoff under the supervision of
Professors Lingxiu Dong and Panos Kouvelis. It is intended as the basis
for class discussion rather than to illustrate either effective or
ineffective handling of an administrative situation.
               BlockBuster Worldwide Stores                                 Blockbuster Revenue


               5000                                          3000


                  0                                          1000
                       1997   1998    1999    2000
         BlockBuster   6049   6381    7153    7677                   1997          1998             1999         2000

                                                                              Total Revenue in Millions
                                                                              Average Revenue Per Store in Thousands
 (1)                                                   (2)

 Figure 1. Blockbuster’s worldwide stores growth. (BBI 2000 Annual Report)
 Figure 2. Blockbuster Total Revenue and Average Revenue per store. (BBI 2000 Annual Report)

        Now add something new, the Internet, and something old, the postal system, and a
new model for movie rentals takes form. Netflix (NFLX), launched in 1998, is the
world’s largest online subscription-based DVD rental service, currently accounting for 3-
5% of all U.S. home video rentals9 but 90% of online DVD rentals. They offer over
15,000 titles to their one million customers12. For $19.95 per month a member can get up
to three titles at a time sent to his house through first class postal service, up to eight for a
higher subscription rate. There are no late fees and no due dates. Members fill out a
rental queue in their online profile and Netflix sends them the first three selections
immediately. When a member finishes with a movie he inserts it in the pre-paid envelope
and drops it in any mailbox; Netflix mails the next movie on the list as soon as they
process the return. Members can rate movies and update their queues as desired, getting
recommendations from Netflix’s CineMatch software. If a member chooses to cancel
service, he has seven days to return his current selections or the retail value of the films
will be charged to the credit card on file. Netflix reports about 7% churn monthly.
Netflix offers only DVD format because shipping of DVDs is 37 cents as opposed to
almost $4 for a VHS4. This somewhat limits the selection Netflix can offer—for example
the original Star Wars trilogy is not yet available on DVD. However, even with this
format limitation, the Netflix library of titles far outstrips the selection of any single
video store.
       Figure 3. Netflix subscription and Revenue growth. Source: Business 2.0 “How Netflix is Fixing Hollywood”

Cost Comparison

Revenue Sharing Contracts: Inventory Costs

For Netflix to effectively compete with Blockbuster in new releases, the fill rate on high
demand “blockbuster” movies must be high. Customers will quickly become frustrated
by a “long wait” status on a new release from Netflix when the local Blockbuster has a
surplus of copies. While Blockbuster owns its DVDs outright, Netflix has established
contracts with most studios wherein Netflix agrees to kickback a percentage of
subscription fees for every movie rented in exchange for the opportunity to purchase
DVDs at cost10. The lower purchase costs allow Netflix to purchase a deeper copy depth
on a title and better meet demand for a title without the substantial capital investment of
full ownership. According to Netflix’s March 6, 2002 SEC filing, page 36:

       After the revenue sharing period expires for a title, the agreements
       generally grant us the right to acquire for a minimal fee a percentage of the
       units for retention or sale by us. The balance of the units are destroyed or
       returned to the originating studio. The principal terms of each agreement
       are similar in nature but are generally unique to each studio. In addition to
       revenue sharing agreements, we also purchase titles from various studios
       and distributors, including Paramount and MGM, and other suppliers,
       including Ingram Entertainment, Inc. and Video Product Distributors, on a
       purchase order basis.

Under the agreements with most of the top studios, Netflix pays on average $1.40 to the
studios each time a new release is sent to a subscriber’s home, a significantly higher cost
than the $1.00 Blockbuster shares with studios for a new release rental on VHS.11 See
Table 1 for more information on revenue sharing agreements.
        According to Fortune Small Business, these costs account for about 20% of
subscription revenues, making the partnerships costly. The contracts generally expire
one year after a film’s “street date”, meaning only new movies are factored into the
revenue sharing costs. Using the CineMatch software, Netflix can guide members to rent
older movies or those released by independent studios, increasing their bottom line and
improving customer service by guiding members toward movies that are more likely to
be in-stock. This leads to the argument that perhaps Netflix should focus more on the
niche market of older, foreign, and independent movies and leave the high demand new
releases to Blockbuster.

Operational Costs: Distribution Centers versus Stores

         Netflix’s distribution system has cost advantages (Table 2). As opposed to the
over 8000 retail locations for Blockbuster, Netflix has just 20 distribution centers across
the nation, with plans to open one or two more each month in 20036 based on the movie
market in that region. According to Reed Hastings, founder and CEO, the company is
able to keep overhead low as the small distribution center facilities have low rent and
require a low number of employees to operate.7 Each is staffed by approximately 12
employees and each processes about 15,000 DVDs per day9. As distribution centers
move into areas, members in close proximity can expect to see turnaround drop from
about one week to just two days, increasing the number of DVDs they can possibly view
in a month. Netflix has experienced a popularity surge in cities with new local
distribution centers. The drawback, however, is that faster turnover and higher viewing
rates result in more postage fees for Netflix, creating a tradeoff between increased
customer satisfaction and increased costs. Also, a typical revenue sharing agreement
requires payouts for each rental of a new release during the first year, so a higher rental
rate will result in more rentals of a film and therein more revenue sharing costs.
Traditional Revenue Sharing Economics for VHS Rentals*:

For the Retailer                         Traditional Pricing         Revenue Sharing
A. Number of Tapes Purchased             10                          30
B. Price Per Tape                        $60                         $9
C. Purchase Cost (AxB)                   $600                        $270
D. Number of Rentals                     300                         500
E. Total Rental Revenue (Dx$4)           1200                        2000
F. Retailer’s Share of Revenue           $1200 (100%)                $1000 (50%)
G. Retailer Profit:                      $600                        $730
H. Profit per Dollar of Inventory        $1.00                       $2.70
For the Supplier                         Traditional Pricing         Revenue Sharing
I. Number of Tapes Purchased             10                          30
J. Price Per Tape                        $60                         $9
K. Revenue From Selling Tapes            $600                        $270
L. Number of Rentals                     300                         500
M. Total Rental Revenue (Dx$4)           1200                        2000
N. Supplier’s Share of Revenue           $0 (0%)                     $1000 (50%)
O. Supplier’s Total Revenue:             $600                        $1270
P. Supplier’s Production Costs I x $10   $100                        $300
Q. Supplier’s Profit                     $500                        $930

Revenue Sharing for DVDs

For the Retailer                         Traditional Pricing (BBI)       Revenue Sharing (NFLX)
A. Number of DVDs Purchased              10                              30
B. Price Per DVD                         $15                             $1
C. Purchase Cost (AxB)                   $150                            $30
D. Number of Rentals**                   300                             500
E. Total Rental Revenue (Dx$4) **        $1200                           $2000
F. Retailer’s Share of Revenue***        $1200 (100%)                    $1000 (50%)
G. Retailer Profit:                      $1050                           $970
H. Profit per Dollar of Inventory        $7.00                           $32.33
For the Supplier                         Traditional Pricing (BBI)       Revenue Sharing (NFLX
I. Number of DVDs Purchased              10                              30
J. Price Per DVD                         $15                             $1
K. Revenue From Selling DVDs             $150                            $30
L. Number of Rentals                     300                             500
M. Total Rental Revenue (Dx$4)           1200                            2000
N. Supplier’s Share of Revenue           $0 (0%)                         $1000 (50%)
O. Supplier’s Total Revenue:             $150                            $1030
P. Supplier’s Production Costs I x $1    $10                             $30
Q. Supplier’s Profit                     $140                            $1000
*VHS Revenue Sharing table first appeared in Cachon and Lariviere, 2001
**Annual Rental Volume estimate is based on 3 tapes per month plan, average use of 5 DVDs per month with
a 25 day month and 3 day shipping. Demand for this release is assumed to be high to moderate.

Table 1: The Financial Effects of Revenue Sharing Agreements
  Figure 4. The logistics of Netflix. (Business 2.0 “How Netflix is Fixing Hollywood”)

           With their current system, each facility fills approximately 98 percent of customer
  orders. Orders that cannot be met by the nearest facility are passed on through time
  zones until they can be filled. An estimated 84 percent of its rental library is available
  within a few days, making turn-around on titles very quick.11 Logistics are clearly an
  important factor in a model such as this. Originally, all returns were checked-in and
  shelved before the new demands were met, making the first half of the day returns, the
  afternoon fulfillment. However, by changing the procedure to simultaneously check in a
  movie and then match it to a new demand the process has streamlined substantially. This
  system modification reduces shelf time on inventory has slowed hiring and reduced labor
  costs by about 15 percent.9 On average about 300 DVDs are unshipped from each
  facility at the end of the day, about 2 percent of the volume that flows through the center
  on an average day, and are stored in a small box at the facility. Each week, any
  consistently unused inventory is returned to the main distribution center in San Jose for
  longer-term storage.4

Operational No. of                      No. of               No. of titles       No. of DVDs
Comparisons locations                   employees            available           available
Blockbuster,          8000+; 1     89,000                    About 1000          Hundreds Per
Inc.                  distribution                           per location,       Location
                      center                                 up to 8,000
Netflix               16           381                       13,500              3.3 Million
                      centers           6            Not available             12,000              Not available

Table 2: Operational comparisons for the top three competitive DVD subscription services.

         As with any successful business idea, Netflix has its imitators. While there are
  many small online companies with a similar product, the two largest direct competitors
  are well known: Wal-Mart and Blockbuster (Table 3).

          In October of 2002 Wal-Mart announced a test program through in
  which customers could rent up to 3 movies for $18.86 per month. Films are delivered via
  the postal service and new selections are sent out as prior selections are returned. Wal-
  Mart currently offers a selection of over 12,000 titles, shipped from its six distribution

          Blockbuster is following suit with its site, allowing up to four
  DVDs at a time for a $19.95 monthly subscription. Titles are limited and all films ship
  from its Arizona distribution center. However, expansion to more distribution centers is
  under consideration. In addition, in July of 2002 Blockbuster started a test market for its
  DVD Subscription Pass program that would allow for members to rent up to two DVDs
  for $19.99/month or 3 for $24.99/month. Members prepay for the service and choose
  from DVDs available at their local Blockbuster store. While this service does not
  provide any benefits in selection or convenience, it does allow for unlimited viewing of a
  DVD. With the enhanced content of DVDs many movie viewers appreciate extra time to
  view the bonus features such as deleted scenes, cast interviews, and behind the scenes
  footage without the late fees. In addition, while Netflix, Wal-Mart, and FilmCaddy
  require at least a day to ship the DVD, Subscription Pass caters to the instant gratification
  market in that members can choose their movie the day that they wish to see it and
  exchange it for a new selection in one transaction. Blockbuster reports that 90% of its
  customers decide on their movie less than 4 hours before making a rental2.

DVD                                    Discs     Number
                          Monthly                                                 Delivery
Subscription               Cost
                                       at a      of Titles        Delivery
Services                               time      Available
  Blockbuster DVD        19.99        2         In Store       Customer          Instant
Subscription Pass aka                           selection      picks up/drops
 DVD Freedom Pass        24.99        3                        off
    Blockbuster’s        $19.95       4         8000           Postal            2-4 days
                         19.95 for    3         13,500         Postal            1-4 days
Wal-Mart DVD Rentals     18.86        3         12,000 +       Postal            2-4 days
    Rent My DVD          $23.95       4         12,000 +       Postal            1-4 days
  Number Slate           $19.95       3         13,000+        Postal            2-4 days
Table 3. Comparable subscription packages offered by the main competitors.
        However, Netflix may be able to defray this direct competition. In June of 2003,
Netflix was granted a patent on their business model for DVD rental. Wal-Mart,
Blockbuster, and any other potential competitors will have to design a model
substantially different from the Netflix model unless Netflix decides to license out the
patent rights. Among over 100 elements of the business model, the patent gives Netflix
intellectual property protection over the way that a customer sets up his or her rental list
and the way the company sends the DVD's.

        However, imitators are not the only competition. Pay-per-view, premium cable,
and Video on Demand seek to serve the in-home on-demand movie market. These
services also serve to the “stay-in” crowd by allowing entertainment selection without
having to leave the house. Pay-per-view and premium cable are available to anyone with
cable, satellite, or digital service; as of 2000 about 75% of households that owned
televisions subscribed to a cable service8. However, cable and pay-per-view are
constrained in their selection to viewers, and the selections are not interactive: they
cannot be paused or replayed and do not offer the bonus features of a DVD. On-line
video rental services, such as Movielink, offer a limited number of films for download to
home computer. Critics of such services say that they are too slow to download and
argue that most people will not want to watch a movie on their computer. Forward-
thinking proponents argue that as the line between home entertainment and computers
continues to blur, and as more homes get broadband, the online video rental services will
gain popularity. Video-on-Demand has attracted a lot of attention, offering a wide
selection of films that can be downloaded to a television set via a set-top box. However,
the technology required for this service is costly and not widely available, limiting the
market. As broadband becomes more prevalent and the cost of set-top boxes decreases,
video on demand is expected to gain ground in the on-demand entertainment market.
Netflix’s Reed Hastings acknowledges the appeal of going digital, but notes that while
every household has postal service, very few have broadband. Also, delivery costs on
downloadable DVD-quality movies can be more than $30, as compared to the 72-cent
roundtrip cost of the current model. But as the costs of digital delivery drop, Hastings
says, “in five to ten years, we’ll have some downloadables as well as DVDs. By having
both, we’ll offer a full service.”10

Customer Relationships

        Anyone that has used traditional video rental agencies knows that the system has
problems. With a strategy known as “managed dissatisfaction”, video stores choose to
stock fewer copies of films than projected demand in the first few weeks following
release. Stocking to meet projected demand would result in huge volumes of excess
inventory after the initial rush. Customers unsuccessfully seeking a specific title
generally leave with a second or third choice, with the stores gambling that the customer
will be disappointed but still willing to come back the next weekend. If a customer does
not have a second choice in mind when he comes in, he is left walking the aisles, an often
frustrating and very time consuming part of the evening.
             Now consider Netflix. With the online personal profile, movie rental history as
     well as the member’s personal ratings of movies is fed into the CineMatch software,
     available free to all subscribing members. A list of recommended movies can be pulled
     up and added to the rental queue with a mouse click. When the time comes to send out a
     new selection, if a particular movie is out of stock, the next movie in the member’s rental
     queue is substituted. This guarantees that while it may not be the top movie on his list, he
     is not settling for just any movie in stock. Netflix claims that they ship more than 90
     percent of all titles from the first three spots of members’ rental queues. However, the
     method by which the high demand movies are allocated is under question. A dissatisfied
     customer, frustrated because most of the movies on his queue were of “long wait” status,
     investigated the process using several accounts13. He tested the same set of 43 movies
     across two types of accounts (3-out and 5-out) and varied the rental rates in the accounts.
     Despite the fact that each account had the same movies in queue, those with lower rental
     volume and thus a higher per-disc revenue for Netflix, consistently had a lower queue
     availability score, meaning more movies were made available to them. In general, he
     found that customers with very high rental rates in recent history are given lower priority
     on high demand movies. (Figure 5) This means that new customers, trial basis
     customers, and those customers who tend to hold onto movies longer and thus rent fewer
     per month are given better customer service. Netflix does not reveal how allocations are
     made. See the appendix for excerpts of his analysis.

1.    Figure 5. Account A paid $30 per month for a 5-out plan, Account B paid $20 per month for a 3-out plan. Lower scores indicate less
      waiting time. For account A in the first shown billing period, the price-per-disc the previous month was about $2, resulting in an average
      availability score of 42 (most movies had at least a short wait). In the second billing cycle, the price per disc history had increased to $5
      per disc and the score improved to 25, meaning about half of the movies had status “available now”. Notice that when account B’s
      average price-per-disc changed from $4 to less than $2 on 4/14 the availability score drastically jumped from an average of 13 to an
      average of 40. Source:
        The CineMatch software also aids in inventory control. Traditional video stores
rarely take the time and money to market smaller films, instead deriving about 80 percent
of rental activity from 200 titles10. Using CineMatch, Netflix can recommend titles based
on a member’s rental history and ratings, regardless of how the film did at the box office.
As a result, 80 percent of rental activity is generated by 2,000 titles, and 97 percent of
Netflix’s titles are rented in a month12. With the decrease in demand on popular new
releases, Netflix can better meet customer demand and saves on payouts for revenue
sharing agreements. The whole experience is catered to a member’s personal tastes and
preferences, making an online transaction seem more personalized than an actual trip to a
video store.

        One customer service drawback to the Netflix arrangement that similar models
will also face is the involvement of a third party, namely the US Postal Service (USPS).
Discs lost en route to the member or back to the distribution center reflect poorly on both
the customer and the service, with both parties likely to blame the USPS for the loss.
Customers with chronic problems have their accounts put on hold, which is frustrating to
a customer that has done everything correctly. Members must file a complaint with the
postal service in order to have their accounts reinstated, a time consuming process during
which they are paying for services that are not active.


       The advent of digital entertainment has changed the face of video rental, and as
technology improves the industry will continue to utilize new developments to better
serve customers seeking convenience, ease of use, and variety all at a low cost. In the
meantime, current models seek a competitive advantage among similar services and
attempt to streamline current operations to make the systems profitable.
Discussion Questions

  1. At what rate should a Netflix customer rent in order to enjoy the value of the
     subscription rather than renting from Blockbuster? How are you going to account
     for the “late rental return” behavior of the customer in your estimate? How
     should the rate be different for different types of movies, new releases, old
     movies, kid’s movies?

  2. Based on your answer to the above question, how would you estimate the variable
     costs incurred by a customer who enjoys the value of the subscription? Can you
     describe two or three customer segments that affect Netflix's margin differently?

  3. In a recent article about the Netflix business model, the following statement was
     made: “Boutique movie studios like IFC films, which released last year’s critical
     hit Y Tu Mama Tambien love Netflix because it provides a huge market for
     movies that can’t muster a widespread theatrical release, as well as for those that
     last only a few weeks on the big screen.” Why do you think it is so? Is after all
     Netflix just a niche market player? Should it become one? Justify your answer.

  4. It looks like success could breed failure. According to a recent story: “But the
     [Netflix’s] spectacular growth is creating unforeseen problems… Customers are
     renting an average of 5.5 movies per month, compared with 4.5 two years ago.”
     Why is this a problem for Netflix? In what ways could Netflix handle the
     operational and revenue sides of this observed trend?

  5. Netflix uses multiple regional DCs while Wal-Mart until recently used one central
     DC for its DVD rental program. Compare the pros and cons of these two
     distribution strategies. What inventory allocation strategy should Netflix employ
     to achieve better usage of inventory? What are the logistics challenges Netflix
     faces at each one of its distribution centers? How should the information on
     inventory be exploited to manage the material flow in Netflix's distribution
     system? What shipping priority to customer rules should Netflix use for limited
     availability movies?

  6. How should Netflix and Blockbuster estimate the life-time value of customers?
     Compare the strategies that Netflix and Blockbuster should use to attract
     customers. How do strategies affect decisions on metrics for customer service

  7. What are the pros and cons of the revenue-sharing contract used by Blockbuster
     on VHS tapes versus its earlier practice? How comfortable do you feel with the
     assumptions in Table 1 in attempting to extrapolate revenue sharing benefits of
     VHS rentals to DVDs and subsequently compare Blockbuster’s and Netflix’s
   benefits? Which assumptions would you challenge? From the Netflix
   perspective, how should the revenue-sharing agreement be structured? Does it
   make sense for Netflix to use revenue sharing arrangements for DVDs in the first
   place? Explain.

8. In a recent article the following information was repeated: Netflix acquired 500
   copies of the Nice Guys Sleep Alone film. They paid only $1 per copy and
   offered a cut of each rental for the first year. The 10,000 rentals of the movie in
   the year allowed the producer to collect $12,000 in rental income. Using this
   information, what changes would you make to your revenue sharing calculations
   for DVDs in Table 1? Explain.

9. Netflix's system collects a lot of information from customers. When ordering
   DVDs for a new release, how can Netflix take advantages of the information in
   making the ordering decision? Is that information valuable to the film studios? In
   what way? What would you suggest to Netflix on using information to manage
   its customer relationship, as well as studio relationship?

10. How do you anticipate the Netflix-Wal-Mart competition to evolve? What is your
    advice to Netflix on what to do with its new patent on DVD subscription services?
    Should they enter licensing agreements with its major competitors? How likely
    do you consider for the implied event in the recent journal article entitled
    “Netflix, a Division of Wal-Mart?” to happen?

11. What aspects and observations of the insightful Michael Porter HBR article
    “Strategy and the Internet” do you see applicable in the video rental setting as of
    today. Mostly Comment on Netflix, Blockbuster, and Wal-Mart supply chain
    models. Is the Netflix business model “disruptive”? What should be the future
    business model for video rental?

12. Blockbuster offers video game rentals. Based on the above analysis, should
    Netflix expand its product offering to game rental? Are there any other product
    categories you might consider appropriate for a “Netflix-like” model? What are
    the product and currently used supply chain features that make them amenable to
    this business model?

  1. Cachon, Gerard P. and Martin A. Lariviere “Turning the Supply Chain into a
      Revenue Chain” Harvard Business Review March 2001
  2. Cohen, Alan “The Great Race” 1 December, 2002
  3. Hyman, Gretchen “Netflix Flicks Switch on DVD Rental Centers” 20 June 2002
  4. Mason, Sarah “Movies on Demand” IIE Solutions, October 2002, p.25-31
  5. Narayanan, V. G. and Lisa Brem. “That’s a Wrap: The Dynamics of the Video
      Rental Industry” Harvard Business School
  6. “Netflix Opens New Shipping Center” 26 February 2003
  7. “Netflix Opens Tempe Distribution Center” The Business Journal of Phoenix 24
      February 2003
  8. “Networks, Party Chairs Differ Sharply on Obligation of TV Networks to Cover
      Conventions” 31 July 2000
  9. Null, Christopher “How Netflix is Fixing Hollywood” July
  10. O’Brian, Jeffrey M. “The Netflix Effect” December 2002
  11. Ostrom, Mary A “With Newer Releases, Netflix Users Can Anticipate a ‘Very
      Long Wait’” The Mercury News 7 July, 2002
  12. (SEC: Netflix, Inc. Form S-1, March 6, 2002, page 36)
  14. Blockbuster Annual Report 2000
         Exhibit 1: Comparison Of Financial Statements Fiscal Year 2002

                                          Blockbuster   Netflix   Industry      S&P 500

Sales Growth (Qtr vs Year ago Qtr) %          7.9       101.30        7.10        -2.80

Income (Qtr Vs Year ago Qtr) %                NA          NA              NA      6.00

Gross Profit Margin %                         76.0        66.2        70.8        47.5

Pre-Tax Margin %                              5.3        -14.3            5.3      7.3

Net Profit Margin %                           3.4        -14.3            4.9      3.7

5 Yr Gross Margin (5-Yr Average) %            74.7        NA          67.8        47.5

5 Yr Pre-Tax Margin (5-Yr Average) %          -1.8        NA          -1.4         9.0

5 Yr Net Profit Margin (5-Yr Average) %       -2.2        NA          -1.8         5.5

Return on Equity %                            4.5         NA              8.6      7.7

Return on Assets %                            3.0        -16.8            5.1      7.2

Return on Capital %                           4.1         NA              7.5      3.6

Income/Employee $                            4,000      -57,000       6,000      11,000

Revenue/Employee $                          117,000     401,000     114,000      287,000

Receivable Turnover %                         33.2        NA          43.2         5.7

Inventory Turnover %                          4.1         NA              3.2      7.9

Asset Turnover %                              0.8         1.8             1.0      0.3
Exhibit 2: Balance Sheet Comparisons: FY 2002 (in millions)

                                   BBI    NFLX
                                   FY2002 FY2002

               Cash and equivalent 152.5    59.8
                       Receivables 184.8    0.0
                       Inventories 452.1    0.0
               Other current assets 169.5   47.3
Total Current Assets               958.9    107.1
Total Non-Current Assets           5284.9   23.4
Total Assets                       6243.8   130.5

                Accounts Payable 757.0      20.4
                 Short-Term Debt 132.8      1.2
         Other Current Liabilities 587.8    18.8
Total Current Liabilities          1477.6   40.4
Total Liabilities                  2076.8   41.2
Total Equity                       4167.0   89.4
Total Liabilities and Equity       6243.8   130.6
Cash Flow
       From Operating Activities 1451.2     40.1
       From Financing Activities -199.2     70.9
       From Investing Activities -1303.5    -67.3
Free Cash Flow                     129.4    13.2
Excerpts from “An Analysis of Netflix's DVD Allocation System”
For the original full version see:


Around January 2003 I started seeing "wait times" on numerous movies in my queue
skyrocket. In particular some movies which were recent, mainstream, and well advertised
releases were impossible to get. About a Boy was one such example. Thus I began a quest
to determine what was going on. It just did not make sense that some of the popular
movies in my queue were so hard to get. I searched USENET but came up empty.

Test Overview

I created a list of 45 movies in my queue that did not have an availability of "Now."
Initially I only added these movies to account "B." I later added these same 45 movies to
three other pre-existing friend and family accounts, which will be referred to as accounts
"C", "D", and "E". Three of these movies eventually shipped to account "A" or "B"
during the period I was collecting data, so in the end there were only 43 movies that were
in common to all five accounts during the entire test period.

Keep in mind that most movies in my queue had a "Now" availability. The following
table summarizes the state of my queue before testing. Future releases have been

                          Number of           Percentage
Availability              Movies              of Total
Now                       214                 73%
Short Wait                48                  16%
Long Wait                 19                  6%
Very Long Wait            12                  4%

This report does not attempt to determine what percentage or class of movies Netflix has
availability problems for. It focuses solely on how movies are allocated when there are
availability problems.

The five test accounts used three different Netflix service centers. A service center is the
place where you are most likely to receive a DVD from and where you return your DVDs
to. Two of the accounts were in the 5 movie out plan ("A" and "C") while the rest were in
the 3 movie out plan ("B", "D", and "E").

An overview of the test accounts:
The Results

To aid in comparing the availability of movies in one queue to another, I created an
"availability score" for each queue. I assigned a weighting to the different levels of

Availability Weighting
Now            0
Short Wait     1
Long Wait      2
Very Long

A queue was then assigned an overall "availability score" by adding up the weighting of
each movie. So lower availability scores mean more movies are available, higher
scores mean more movies are unavailable. To put it another way: low score good, high
score bad!

The following table summarizes the availability stats for each test account:
The number of rentals in the previous billing period is charted against average
availability score for the current billing period below. Accounts in the 3 movie plan are
shown in orange while those in the 5 movie plan are shown in blue.

In version 1.0 of this report I indicated that I thought the rental plan you are in is not
affecting the availability score. I.e. if customer A who is on a 5 disc plan and customer B
who is on a 3 disc plan had the same rental habits over a period of time, they would have
the same availability rating.

A slashdot poster suggested average rental price was in fact what is affecting availability,
and this it would take into account the various rental plans. Or at least in the unlimited
plans. The following chart shows per-disc rental cost charted against availability score. I
have charted two data sets: the previous billing period average disc cost and the average
disc cost over the last two billing periods.
The points stuck on the X axis are for "B" during the trial and 1st month. The clustering
of both 3 and 5 disc plan samples in the same area does indicate that your rental plan is
part of the equation! This is great news.

Given the very small data set, it is difficult to tell whether Netflix simply places renters
into different priority "buckets" or fine tunes the algorithm on a per-customer basis. Note
that as of the 1st quarter of 2003, on average each customer rents 5.5 discs per month. I
do not know whether this is an average across all plans or only the $19.95 plan.
Assuming a $19.95 plan, this translates to $3.63 per disc. This would put the average
customer in the second cluster from the top above.

Shared By:
wu yunyi wu yunyi