Film Tracking or How I Learned to Stop Worrying and Love Opening Weekends

How do we track movie performance?

In a late summer Deadline article, Anita Busch wrote about the recent poor performance of movie tracking services and laments the studios’ wasted money spent on wildly inaccurate box office projections. Ms. Busch is not alone in highlighting this issue – in recent months reporters for many industry publications have written articles bemoaning the state of the so-called “predictive reports.” In her article, Anita cites an industry insider who thinks “[the tracking services] need to go a little deeper in their research, in the way that they ask and who they ask.” According to the article, some blame the tracking services for not being able to keep up with the habits and influencers of a generation of consumers making nearly constant use of smart phones and social networks in deciding how to spend their time. While the commonly used tracking services have used focus groups to understand consumer behavior for decades, they’re less savvy at taking into account online behavior. Ms. Busch notes that the focus of tracking companies should be on accurately representing projects’ profitability, since that’s what most stakeholders are concerned with after all. She closes the article with a question: “So how do you fix it? Anyone?”

[I am eagerly raising my hand – this slows down my typing but I’m hoping sooner or later they’ll call on me]

Focusing on overall profitability sounds like an excellent idea! But before deciding how to fix a system that may or may not be broken, let’s take a moment to consider what these companies are trying to do and how they arise at their projections…

The purpose of these services’ reporting is to provide a snapshot of the marketplace and how specific marketing materials are being received at the current time. They attempt to do this primarily by polling groups of likely moviegoers (on phone, web and in person), asking whether they’re "aware" of specific movies and to report on the likelihood they will see it. These respondents are broken into quadrants: males <25, males >25, females <25, and females >25, with racial divisors sometimes incorporated. These are relatively broad dividing factors, but as such they serve as flexible indicators of awareness of many types of films and film releases. In addition to weekly reports, many of these firms offer bespoke services like testing trailers, TV spots and print ads, conducting exit polls, hosting recruited-audience screenings, organizing focus groups, testing titles and positioning and studying franchises. However, it is the weekly reports that are purchased in large numbers and used to project opening weekend returns – projections responsible for the recent bad press when widely off the mark.

Boiling it down, these weekly tracking reports are trying to determine 2 things: (1) the number of people aware of upcoming movies; and (2) how many of them intend to buy tickets at the theater. Using polling to study the success of a film’s marketing campaign probably feels worth the investment to most of the studios and production companies that subscribe to it. However, the questions are whether these reports should be used as predictive indicators and whether they, along with actual box office reporting, need be the only metric for film investors to value types of films.

Those of us who take an interest in such things want a larger selection of touchpoints to gauge film performance. We want to be able to place films in a clear context so not every film is being judged on the same scale. Investors in small films don’t expect anywhere near the level of return that a studio blockbuster commands. Yet, in broad strokes, the industry and media look at next weekend’s box office as though all the films are running in the same race. It’s wrong to act like we can gauge all films against one another in a single list. But more than that, this lack of sophisticated reporting and predictive metrics is detrimental to the film industry by making it appear even more volatile than it actually is.

When do we track movie performance?

These reports highlight the industry obsession over the all-important opening weekend. Why do we put so much emphasis on a film’s performance during opening weekend? Because we do, that’s why. Sure, one can examine how over time, the industry has matured in a way that it has naturally come to focus on opening weekend because it’s a relatively straightforward way for distributors and exhibitors to cull poorer performing films from a finite number of movie theaters. There’s nothing wrong with that per se – the industry has to create some Darwinian mechanism to allow specific films to survive and perish in the theatrical window and opening weekend has risen to become a key indicator. But does such an overwhelming focus on the first three days actually make sense in a world where a film’s revenue is earned over such a longer period and through so many other avenues than just theatrical exhibition?

The book publishing industry, which itself is in the midst of some much needed renovation, uses the term frontlist to mean current titles – books that were released recently and which are still receiving conscious marketing resources. Backlist refers to anything that doesn’t fit that category. While many publishers do focus on frontlist to drive the bulk of their revenue, an increasing number of companies rely equally if not more so on maintaining a backlist that performs strongly for years. By achieving stable sales from a sizable backlist library, these companies feel comfortable when taking measured risks on the frontlist they believe in rather than nervous over the success or failure of each new title.

The analogy between film and publishing isn’t exact, but if you were to draw a line between film and publishing release cycles, frontlist would be equivalent to the theatrical window (assuming a typical release pattern). Backlist would be the windows after a film is in theaters (VOD, DVD, etc.). Studios don’t need to be taught the lesson of how important “backlist” revenue is to their bottom line. The money made from film libraries dating back nearly a century continue to bolster studio revenues and require much less servicing than the heavily marketed premiere of their next film. However, in part because of the sheer size of many film budgets, they continue to place enormous emphasis on the film that’s premiering rather than focus on extending the active lifespan of past releases.

We do sometimes read articles reporting on aggregate film sales, especially as films cross their next $100mm threshold or meet some other notable milestone. However, the focus on a film’s performance wane week over week as it ages. For the consumer or hobbyist reader, this is likely the way it should be – popular film journalism should be entertaining so it should focus on what is new. However, film investors need the ability to continually view a film in its proper context, throughout its life, to understand how it’s performing compared to expectations, what those expectations were, how other correlated films are performing at that time, etc.

The film finance index is being designed to provide just that type of context. Not to replace tracking services. Not to diminish the importance of opening weekends. But to add a tool to an investor’s toolbox that provides clarity about the performance metrics they’re seeing. In my next blog post I’ll delve into my latest progress toward this end.

Revenues of Future Past

the-numbers-banner

I’ve been speaking with folks at two companies with very different approaches toward using and creating film revenue data. Both are important and insightful in different ways. The Numbers is a website combining many of the variables I’m taking into account with the film index. It has catalogued over 18,000 films, recording things like budget, release date, attachments, rating, genre and budget. It then goes on to report on their various revenue streams, including Box Office and DVD. While less reported on through the consumer-facing site, the underlying layer of data, named OpusData, also has some information about VOD sales.

While Box Office is widely available and reported on by many sources, DVD and VOD are much more elusive, so I was particularly interested in this part of the reporting. It appears the information about ancillary revenue is a combination of historical information and projections based on a survey of historicals and other factors. One service OpusData supplies is bespoke projections for production companies, distributors, researchers – basically anyone who has a focused interest in making an educated projection about an upcoming film project. Over years of classifying films along a fairly complex rubric and having access to real information regarding ancillary revenue, the system has learned enough about this to begin incorporating revenue projections beyond just box office in some cases. Perhaps even more exciting, the service also collects information and makes projections related to P&A spend, one of the pieces of data most difficult to find and validate in the film world. I am in the process of evaluating what OpusData offers and assuming I gain confidence in its validity and usefulness, I may incorporate these much-needed data into the index.

1973419

The Hollywood Stock Exchange (HSX) takes a very different approach to projecting film returns.  Described as a game, the system uses a community-based approach to form an idea of how well or poorly specific films are expected to perform. Participants can buy and sell “shares” in everything from films, to actors, to opening weekends. As with most exchanges, betting against projects that one thinks is overvalued (commonly called “shorting”) is also allowed.

Depending on how widespread participation in this game becomes, one could make the same arguments about its efficiency as a financial exchange in which real money changes hands. It is important to understand the amount of bias embedded in the system, by virtue of the fact that shares are being “bought” and “sold” without any actual financial consequence to the trader. This can likely be solved for fairly simply. With enough participation by rational “investors”, something like HSX could end up projecting film revenue as well as if not more efficiently than other systems, though all of the underlying motivations for specific valuations may not be fully transparent. I am continuing to explore HSX’s data to see if there might be an opportunity to incorporate it into the information contributing to the index. In the meantime, it's been fun creating a portfolio of films and hone my summer blockbuster intuition.

A look at existing film performance research

school head icon

Forest for the trees: Risk categorization criteria

Over the last month I’ve been wandering through a forest of data - exploring and compiling various pieces of information related to 1350 films that received a US theatrical release in 2012 and 2013. This data will be used to train the initial rubrics that will guide my separation of films into four risk levels.  While I’m still in the process of finding and organizing this data, I wanted to reach out to readers to ask for your thoughts on my approach. As a reminder, I’m initially dividing the index into 4 sub-indices, each with a distinct level of risk. See my last post for more detail. My job, therefore, is to properly select criteria that correlate to risk (in terms of films’ financial returns). The 3 criteria I believe will be most important follow:

Marketing budget

I believe this is one of the most important factors contributing to a film’s success or lack thereof. You can tell a unique story through an amazing, well-crafted film, but if you don’t properly get the word out about it, it won’t be seen by many people. Unfortunately, I’m discovering that accurate information about films’ marketing budgets is quite difficult to find.

I’m planning to compile information from filmmakers, distributors, ad agencies and journalists to triangulate/approximate marketing budgets for specific titles. However, this may not be fruitful in a lot of cases. And it’s hardly a scalable approach. So, dear reader, where else might I look for information about marketing budgets? Extra points for accurate, scalable sources.

Production budget

A film’s budget impacts everything about it. If a story hasn’t been sufficiently supported through its budget, an amazing story and/or cast won’t shine through. Nor will a marketing budget be as effective, regardless of its size. This is not to say that all films have to have a large budget – many very low budget films have still performed very well, both critically and financially. While I believe it highly likely there will be strong correlation between film budget and ability to drive revenue, I also believe films with distinct profiles have distinct budget levels that will indicate propensity for success at the box office and beyond.

Sources like BoxOfficeMojo are fantastic aggregators of data like film budget. However, it’s not assured the recorded budgets are correct. Additionally, I’ve found sites like this don’t record budget for hundreds of low budget indies. Where might I find budgets for these smaller films? How might I confirm the accuracy of reports made by data aggregators like this?

Attachments

Films are not commodities. Films’ casts, directors and producers undoubtedly have enormous impact on the finished product. Therefore, I’m positing that attachments themselves have significant impact on the potential performance of a film independent of its production budget or marketing budget.

The creative players in a film are some of the easiest pieces of information to discover with certainty. However, unlike budgets, attachments are not numerical in nature. To approximate a numerical system, I’m taking on the highly imperfect task of devising a scoring system within which to place key attachments, based on the financial success of films in which they previously participated. I am trying to create a science in something that is obviously unscientific. Any opinions about factors I might want to consider along the way are highly encouraged.

I should note that I will be tracking other sorts of data as I organize my training data. Critera like distribution strategy (wide, platform, limited, day-and-date, ultra-VOD, etc.), source (studio, mini-major, established indie filmmaker, unknown, etc.) and origin (existing blockbuster book, original material, etc.) are likely related to level of financial risk. However, these may end up being largely duplicative of the above, more complex criteria. Also, I’m limiting the data pool at this early point in hopes of not muddling other correlations that may prove strong.

I’m lucky to have had interested mathematicians and data scientists reach out to help me discover and organize meaningful conclusions from my raw data. Engaging with folks who read these posts has been incredibly helpful. I want to talk with even more people – so, please, comment on this post or email me directly at colinwhitlow@gmail.com if you’d prefer. I’m open to all ideas and questions.

Risk, Normalization and Building a Sandbox

Constructing a useful film index amidst a constant fluctuation of industry conditions feels a bit like drawing a box in the sand on a windy beach. Any grains of data that fall within the box would be included in the index because of their perceived usefulness toward projecting future performance. However, with current viewing habits and production and distribution practices continually shifting, the outline of that box continues to transform. I’m currently concentrating on two key challenges in defining the boundaries of the sandbox: normalizing the index scoring system and distinguishing films according to their level of risk.

Normalization of the index score

Simply looking at box office performance at a given point in time tells you very little. It tells you something about the strength of titles currently in theaters. It tells you a lot about the importance of the release cyclicality (i.e., Thanksgiving and July 4 weekends are huge overall ticket sellers), which then becomes a self-fulfilling prophecy for distributors planning their release timing. Film investors fully expect to see ticket sales spike in at certain times because that’s what they’ve consistently done historically. But they need more context to begin to understand more about performance of specific titles and categories.

Scoring this year’s data against the same week from the year prior begins to give a little more information. I’m of the belief that comparing the current year to anything further back than the previous year would begin to warp the conclusions one might draw. For instance, consider how much Netflix, HBO Go, Hulu and other platforms have shifted viewing patterns over the last two years. To blindly weigh 2013’s July box office performance against 2005’s would be to compare two very different environments and would confuse or mislead any investor trying to draw meaningful conclusions.

Many other forms of normalization will likely need to be employed as I develop the index’s underlying formula – analysis of the statistical significance and reliability of each type of data available will help lead toward meaningful normalization factors. For instance, I might account for inflation and/or remove 3D titles from consideration. Eventually, a model might be able to account for more nuanced factors like macro weather patterns and events. In the end, simply comparing one year to another provides limited information for many reasons. Release patterns are in constant motion. Aside from several standard big weekends, which are likely to remain bellwether box office indicators, releasing a specific film during a specific week tends to be based on a host of considerations unique to that period. Not to mention the fact that holidays don’t appear as the same “week” each year. When viewed as a graph, this calendar variation will create peaks and valleys that don’t make sense in comparing two years. Such a comparison also doesn’t speak to the actual volume of revenue earned. Clearly, while scoring one year against another can provide certain information, it is only one tool in the toolbox.

BO Compare

To the left is a basic representation of 2013’s weekly box office as compared to 2012’s. Any positive percentage indicates 2013 performed better that week than the same week in 2012.

* For those curious about the pre-Thanksgiving dip, 2012’s opening of Breaking Dawn, Part 2 dwarfed anything 2013 had at the box office that week.

Risk categories

A common factor used to categorize various investments is level of risk. Risk level is a core consideration for investors, who need to be able to compare other holdings in their portfolio with the one under consideration and weigh the entire basket against their overall risk tolerance. Film investors attempt to identify these same risk characteristics as they relate to specific projects. Sophisticated film investors attempt to create diverse portfolios to try to ensure several flops don’t overshadow the hits. A film portfolio, typically referred to as a slate, is not a new idea. However, successful film slates are not nearly as common as lauded mutual funds because visibility into specific films’ actual level of risk is typically not great.

Rather than getting bogged down with limitations of identifying real projects to make up a film slate, I want to first describe the categories I might employ, using stock category definitions as a model.

- The least risky category of stocks is often referred to as “income stocks.” These are typically thought of as slightly higher risk and return potential than bonds. Utility stocks are in this category, since investors have faith that people will continue to need things like electricity and water.

- “Value stocks” are those that have historically been a good investment and which are thought to be undervalued at their current price. The idea is that buying undervalued shares allows you to enjoy additional upside as the price returns to its proper level.

- Stocks with strong historical and projected growth rates are categorized “growth stocks.” Investors expect a strong return on equity with these. Many technology and alternative energy stocks are placed in this category, assuming they’re at a scale offering reasonably low volatility and established enough to have historical data to reflect a trend of high growth.

- Then there are riskier types of investments, including smaller companies with no proven track record or significant market presence and companies from emerging markets. These investments are exposed to a higher than average level of volatility due to lack of liquidity, political and currency issues that might disproportionately affect them and generally uneven growth, often due to their small scale.

Sandboxes

I believe there are some analogies to be drawn between risk categorization in films and stocks - translating these definitions may help me in drawing lines around films for my own index. Once the boundaries themselves are clearly defined, determining the content to be included within becomes much easier.

The Clarity and Color of Film Finance

“Let me show you how this is done. First thing, hold the glass up and examine the wine against the light. You're looking for color and clarity. Just, get a sense of it.” – Miles Raymond, Sideways

I believe the film industry is weakened by the absence of an objective, transparent measuring tool for would-be film investors. Many mature industries have created an index of some sort to help the investment community get a read on the sector’s financial climate. The Dow Jones Industrial Average serves as a benchmark for mainstream market activity. The Case-Shiller Indices are well-respected indicators of real estate performance. But the film industry has no such barometer. Some risk prone investors become involved in films with virtually no useful financial information to go on, making a hopeful jump into the abyss with the belief that the journey will have been fun regardless of its outcome. A small group of experienced investors understand how to create assurances for themselves, increasing their chances of financial success. The overwhelming majority of investors, however, choose not to invest in film at all, finding alternative vehicles that provide greater transparency into the mechanisms related to financial returns and the set of past comparables. Both filmmakers and investors suffer from the continued opacity surrounding investment opportunities.

Under the belief that the presence of an objective film index might help alleviate this issue and lead to more sustainable and intentional investment practices, I am researching methodologies through which such an index could be made. Once I have collected sufficient information about the challenges, data types and partners that would need to be involved in making such an index a reality, I will move toward the execution stage. At the moment, I find myself defining the rules for this process…

First thing’s first – what is an index really? In simple language, it’s a set of investments combined to form a score that can be compared at various places in time relative to its own base value.

Line graphs depicting stock performances over time are fairly standard visual indicators commonly used to aid the decision-making process of investors. However, there is no simple or obvious way to make a similar tool related to the performance of film investment – at least not without setting a few things straight. There are a plethora of ways in which films are unlike stocks - below are a few:

comparison

• - Stocks hold value that is theoretically ongoing whereas the revenue-generating life of a film is finite. A film index needs a standard basis to gather revenue information.

• - Companies available for trade on the public market are highly regulated while even studios and production companies owned by publicly traded conglomerates do not disclose financial returns on a per-film basis. A useful film index requires data its users trust.

• - Stock markets allow for high volumes of ongoing trading from many different sources, which all fall into a standardized system, whereas films typically require funding at the front end from a small number of sources and generate revenue from a widely diverse and disparate set of streams during specific windows.

Luckily, there exist other indices related to investments I find much more analogous to films. One such index is Liv-ex, a global marketplace for buyers and sellers of fine wine. As I’ve begun studying the Liv-ex system, I’ve found the practical and philosophical questions its founders tackled highly analogous to the exercise I must go through in constructing the film index. Two core elements are of utmost importance:

1. Defining the criteria for inclusion

As Liv-ex’s director, James Miles, said, “To be regarded as a fine wine, a wine must have the potential to both improve in bottle and appreciate in value, and be actively sought after in the secondary market. To satisfy this requirement, a wine must have a long track record (often centuries rather than decades) and have received strong critical acclaim.”

It is essential that a film index articulate a fine level of specificity in setting the boundaries that demarcate what films it includes. There are many different types of boundaries one might choose. The key, I believe, will be to select a rubric that is meaningful for investors and filmmakers alike and that produces a score that can be used to make projections beyond the list of included films into other parts of the industry. Is an ongoing slate of films made by 25 film producers meaningful? What about a set of sub-indices built within genres? Should the budget of the film be a factor? The length? With the aid of some statistical analysis, practical considerations regarding the ability to get specific data and common sense, I will gradually determine the pieces that might best be used to construct the model.

2. Defining the index’s value

In Liv-ex’s case, James Miles points to the index as a tool investors can use to calibrate fine wine as an asset class within their portfolio, comparable to others such as gold and real estate. He points to the ability for “growers to finance their crop, which for fine wine is often not ready for drinking for a decade or more” – essentially the idea that it provides greater stability for the content, errrr, crop producers. And he points to the burdens of transaction fees leading to low liquidity and slow working capital cycle times throughout the industry, with the transparency of an index serving to lessen those challenges. The analogous value of an index to the film industry is fairly direct.

Screen Shot 2014-02-25 at 9.53.42 PM

"Liv-ex Fine Wine 1000" Index

I have begun speaking with Liv-ex staff and will continue to seek their guidance over the coming months (as well as from other indices serving similarly atypical financial instruments). Down the road when I begin to experiment with functional models, my initial attempts will almost certainly need refining. Everything from the Dow to Liv-ex have gone through periods of revision. With each iteration I’ll seek to bring a useful model for insight into film investment into greater focus. I’m looking forward to working with many of you and sharing my progress.

* To learn more about Liv-ex, visit their website here.