Appendix X: Release Sequencing Theatrical Release Timing Peak audiences (X-mas; Thanksgiving, Summer etc.) Peak attention (uncrowded d period) summer movie season is mainly a US phenomenon Release Timing Trade-off between capturing as much of the revenues during the main season as possible while avoiding the competition, which is trying to do the same. Source: Journal of Marketing Research, 1998 1
But theatrical revenues are declining in importance for film companies Release Sequencing Windowing Total Spending on Filmed Entertainment ($ Millions) Year Box Office Home Video Television Total 1990 $5,022 $11,107 $8,245 $24,374 2001 7,622 24,906 15,611 48,139 2002 7,948 26,168 16,589 50,705 Basic Principle for Release Sequence Strategy First, distribute to the market that generates the highest marginal revenue over the least amount of time Then, cascade in the order of marginal-revenue contribution Source: Veronis Suhler; The Publishing & Media Group, Mtion Picture Association of America, Adams Media Research, Video Software Dealers Association, Consumer Electronics Association 2
Sequencing of Distribution Domestic theatrical Foreign theatrical Pay-per-view Worldwide home video Pay TV Foreign TV Network TV Syndication TV Impact of Piracy: Compression of Release Sequence The Perfect Storm, opened internationally a few days ahead of its U.S. opening. Video rental sooner now But early release reduces help distribution of favorite flow of press stories Marketing campaigns have directors and stars visit each country as their film is Source: The Hollywood Reporter, October 10-16, 2000 about to released. This becomes harder. Video Release Sequencing Model Knowledge of the sales parameters in the first channel (theaters) helps predict sales in the second channel (video rentals) and help strategy for second channel Q: When to release film in 2 nd channel? t * 2 = 1 m v ( ) 2 2 B nm 1 M ln v 2v1 M + v 2 n The optimal time to release a movie depends on p, and a joint optimization problem involving both price and time. Assume that price is set exogenously. In calculations, article authors set p = $63.00 and r = $2.50. Source: Donald R. Lehmann and Charles B, Weinberg, Journal of Marketing, July 2000. v T p r Source: Donald R. Lehmann and Charles B, Weinberg, Journal of Marketing, July 2000. 3
Results show that the optimal time to release a home video depends on the opening strength and the decay rate of a movie. Those parameters can be determined early enough in a movie s run to make the video release timing decision. Source: Donald R. Lehmann and Charles B, Weinberg, Journal of Marketing, July 2000. Conclusion on Vertical Integration of Production and Distribution Double-edged sword Synergies has been exaggerated by empire-builders and deal brokers Vertical integration has also disadvantages Key factor is attractive programming Control of release sequencing and secondary distribution can be achieved by contract, and do not require ownership What is the movie release sequence? Jaws Reinvents Hollywood s Release Method 4
Prior to 1975, Hollywood used a platformed method of releasing its movies. This meant movies were originally i released in select theaters and added more theaters in following weeks and months Deciding to forgo traditional mold, Universal Studios released Jaws on more than 400 screens nationwide, the biggest release up to that point. It also had one of the biggest nation-wide prime time ad campaigns in its day. These two factors helped make it wildly successful; it eventually grossed $260 million Hollywood believed it had a new formula for building a blockbuster. Three decades later, the Jaws formula remains Blockbusters are released on more than 7,000 screens http://www.the-numbers.com/movies/1975/0jws.html 5
However, a characteristic that distinguish most of today s blockbusters from Jaws is the fact that they gross large amounts on opening weekend, with revenue plummeting in following weeks. X2 earned $85.6 million in its opening weekend, but couldn t manage to record half that amount the following week The Hulk brought in $62 million its first weekend only to see its numbers drop 70% the subsequent week http://www.the-numbers.com/movies/2003/hulk.php The Jaws model does not guarantee a hit. However, what can be inferred is that if a movie is marketed well and is released on thousands of screens, the chances of a big opening weekend are high. Most studios spend their bulk of their marketing budgets in the weeks prior to a films opening because of its earning patterns. In 1993, the top ten movies made half their total gross in the first three weeks. 6
In 2003, the top ten movies made three quarters of their total gross in those weeks. Studios are all to happy with this trend. They take in 90% of the opening week s take, with the rest going to the theaters that screen it. As the weeks progress, studios get an increasingly smaller percentage of the box office take. Models to Predict Box Office Success Various models constructed to help predict audience responses 7
BOXMOD Approach One approach involves the analysis of a consumer s decision to see a movie in two steps First, person decides to see the movie Second, person goes to a theater and views the movie Based on these analyses, the studios can predict a movie s success based on indicators such as rating or genre Competition BOXMOD Use formulas to predict success of movie at the box office Measures the impact of competition on box office performance Bayesian Model Utility of a movie: M M it ot Uit e I = O + e t = O t j O + j it U jt t U jt e I I jt jt Outside good of a movie: 8
Mit: proportion of people watching movie i in week t Mot: proportion of people not watching any movie in week t Uit: utility of seeing movie i in week t Iit: indicates whether movie i is in theaters in week t Ot: utility of outside good in week t Another approach (Steven Shugan) involves analyzing the team of individuals that created the movie (directors, actors, producers) Goal of this approach is to predict a movie s success early on in the production process Can take the results to identify the probable box office results based on previous success of production team. Beneficial fiilsince analyzing only one of the factors such as an actor is not enough to predict potential box office results However, a studio can have a highly successful movie with an actor but isn t guaranteed to reproduce those results with the same actor if a different combination of producers and writers is used. 9
The Markov Chain Model involves the use of word-ofmouth to predict success Use marketing and reaction of a test audience at a pre-screening to analyze potential success Another model involves studying the allocation of multiplex screens to maximize profitability Difficult to have a universal predictive model for a number of factors: Lifecycle of Movies Potential of Seasonality in Sales Patterns Continuity of Product Lifecycles The lifecycle of movies are usually very short with sales peaking during the first week of release 10
Sales Patterns While most movie sales peak during the first week with rapid declines in subsequent weeks, this is not always the case The blockbuster type movies (64% of movies) have the rapidly declining sales pattern But for sleeper type movies (36% of movies) sales increase gradually and peak in between the third and sixth week of release Seasonality There are established good (July 4 th weekend) and bad weeks (middle of September) Movie studios vie for ownership of these good weeks by signaling to others which week they plan to release their blockbusters However, exceptions: Sony released its 2002 blockbuster, Spider-Man, a week before Labor Day (viewed as a bd bad week) 11
By the end of its second full week of release, the film had grossed $285 million. Thus disproving the good week/bad week model, and that competing for peak viewing weekends may be counterproductive http://www.the-numbers.com/movies/2002/spidr.html http://www.the-numbers.com/movies/2002/spidr.html Continuity Unlike consumer products that are successful largely due to consistent quality, movies are successful because of differentiation Consequently, it is difficult for movie studios to construct a predictive model since continuity, a major factor in the success of consumer products, is its ability to produce the same results 12
Simultaneous Release 2929 Productions Bubble broke the traditional release window for the film industry Challenging the traditional release window model Directed by Academy Award winning director, Steven Soderbergh, Bubble is forgoing the traditional release window model and releasing it simultaneously in theaters, cable TV, and DVD. Wagner and Cuban, the owners of 2929, financed the production, and control all the distribution channels. 2929 Productions owns Landmark Theaters (215 screens in 15 states) And the HDNet cable network. 2929 also distributes its DVDs. 13
Owning the entire production and distribution chain end-toend has been an overarching goal for 2929 and is how a simultaneous release possible. National theater chains refuse to screen a film that could undercut their own business by being distributed on DVD and cable at same time. 2929 believes that the total audience will be larger under the new model, since many people who wouldn t go to theater would generally buy the DVD. (Which retails at $30) preserves incentive to go to theater first Small revenue sharing (1% of DVD revenues for Bubble will be given to theater owners who show the movie). 14
Movie will be marketed under one campaign rather than 3 under the traditional model. This helps small-budget projects. A third benefit to simultaneous releases: less demand for pirated advance copies if you can buy the actual ones when you want them. 15