Film festivals and market intelligence: From audience surveys to data analytics?
Audience research has become an increasingly significant aspect of film festivals. The development of hybrid and online film festivals has enabled opportunities to generate more knowledge about the festival audience. That resembles broader developments in today’s culture and society, which relies on online data to take audience knowledge to a higher level. Indeed, industry observers and academics call this the era of datafication, in which audience intelligence is more than ever believed to impact on industry practices and strategies.[1] This article explores how film festivals engage with online data to enhance audience knowledge.
In the film industry, powerful video-on-demand (VOD) platforms such as Netflix and Amazon rely heavily on online data intelligence gathered from audience members.[2] They have developed a knowledge system that revolves around big data, algorithm technology, and individual consumer preferences. While engagement with online audience data is often described in relation to the most powerful VOD platforms, there are benefits for the exhibition sector as a whole. Film festivals, but also theatrical cinema exhibitors, can acquire a better understanding of their audience base, particularly with the development of hybrid and online screenings for films. They do not present themselves as technology-driven companies, and they do not rely on data in the same way as the likes of Netflix and Amazon, but they engage with audience data in one way or another.
For film festivals, data opportunities raise all sorts of questions about the scale and depth of audience research. What interest is there amongst film festivals to engage with more data? What capacity do film festivals have to undertake audience research? And, given discussions about the controversy around technology-based intelligence, to what extent do they want to rely on online audience data? Based on such questions, I want to bring about a sense of awareness and realisation of film festival engagement with audience data. While audience data begins to play a more significant role in film and media industries, we need to know what purposes it could serve for film festivals.
Research methods
Research undertaken by film festivals is often based on audience surveys. Most film festivals use survey findings to produce annual reports for stakeholders, such as public funding agencies, corporate sponsors, and individual sponsors. As Ragan Rhyne already described in 2009, data and insight reports are amongst the administrative tasks of film festivals that have taken more relevance over time (my emphasis).[3] And with the rapid development of digital technology in the past 15 years or so, reporting about film festival performance (including the audience) is increasingly based on quantitative data analysis.
Audience surveys are relatively easy to produce and analyse, and they provide a baseline understanding about festival audiences. They are often designed to collect information about audience demographics, their interest in the film festival, the films they watch, and since recently their opinion about hybrid and online festival formats. While surveys may be carried out during in-person festival gatherings and/or online, the downside is that such research can be time-consuming, particularly in terms of finding respondents for a representative sample of audience members. They can be undertaken on a small scale with hundreds of respondents, or on a large scale with thousands of respondents. Large-scale surveys require extra effort, but the results are often more precise.
What has changed with the development of online screenings is that film festivals can take advantage of streaming platforms with built-in functionalities. Examples of such functionalities include the quality of video streaming, anti-piracy protection, platform templates and visibility, information about individual films, payment methods and revenue reporting, and of course data analytics about the online festival and their online audience. Such data is usually automatically generated within the limits of the capacity of data analytics programmes. It can be divided in various sections, such as sales reports for film festivals and audience demographics.[4] For individual films, it can provide information about the number of audience views, when they were watched, from where, and for how long. Such information is representative for online festivals because they are specifically organised for an online audience. For hybrid festivals, however, it is representative for a part of their audience because they are organised for physical and online audiences. But also for them, information about their online audience could be valuable to draw comparisons with their physical audience – for example, in terms of the extent to which they reach younger and older audience members.
In addition to screenings of individual films, festival’s organise events that provide context about films, such as introductions to films by filmmakers or festival directors, and discussion sessions with audiences. They often become available as videos on popular platforms such as YouTube or Vimeo, which also provide access to automatically generated data about audiences. Some of that data overlaps with the data collected by film festival platforms.
There are more factors related to data analytics that should be considered, particularly in terms of reliability. One argument is that data from online screenings is reliable because it is based on raw data generated by technology. But that raw data is generated by the set-up of standards and measurements in data analytics programmes. Because external providers often design platforms for online film festival screenings, it is worth mentioning that they also exercise control over the process of analysing data, either by themselves or through data intermediary companies.[5]
The online film festival audience
This section provides several practical examples of audience data collected by online film festivals. I conducted interviews with Thessaloniki International Film Festival, Göteborg Film Festival, and Dublin International Film Festival.[6] Based on insights from the Thessaloniki online festival edition in November 2020, the Göteborg online festival edition in February 2021, and the Dublin online festival edition in March 2021, I identified several themes that inform their engagement with audience data.[7]
The first theme describes their engagement with surveys and automated data analytics. I use the term automatic data here to refer to automatically generated data, which is collected and analysed by data analytics programmes. What is interesting about surveys and automated data analytics is that one method does not exclude the other. Film festivals recognise the value of both methods. As Managing Director Mirja Wester (Göteborg) notes:
You can get so much data now: for example, when people started the film, when they got tired, and what they liked. But we used the data from the system, and we did a large audience survey.
While Wester notes that new data opportunities could complement audience surveys, automated data is still a relatively new source of intelligence. Clearly, the value and validity of online data needs to be proven over a period of several years. In the case of Göteborg, they organised an online film festival in February 2021 and a hybrid children festival in October 2020, which provided another opportunity to gather knowledge about online audiences.
The second theme identified from interviews is the geographical spread of audiences. Hybrid and online film festival editions have revived discussions about audience reach, with arguments that festivals were able to reach a nation-wide audience, well beyond the city or region from which they were organised. As festival programmer Gráinne Humphreys (Dublin) notes:
Our festival has always been very Dublin centric, to the point where we would have very few people coming in from surrounding counties. So, it is genuinely interesting to see that it actually has that reach. And for the small amount of industry events that we did, that there was a reach beyond as well.
On the other hand, there are also reservations about reliance on geographical spread because of the continuing function of festivals as community events. Jalladeau (Thessaloniki) explains that is important to find a balance:
Geographically, the festival is expanding to the whole country in a way. The hybrid character of the festival allows us to reach a nation-wide audience. But we keep in mind that we are a community event. The roots of the festival is to be together and there is no way we neglect that, because our ambition is not to be a platform.
The third theme that came forward in interviews is more specifically based on audience consumption. Dublin worked with the external platform provider Eventive, through which they collected data about the films that individual audience members watched, how many films they watched, and when they watched them. That was particularly relevant to analyse how audiences engaged with ‘live’ or ‘timed’ screenings alongside on-demand viewing opportunities. One finding, as Humphreys (Dublin) notes, is that some audience members followed a timed schedule, but they also developed their own additional viewing schedule:
Our audience could experience the online festival as usual, meaning that they had a screening at two o’clock, at four o’clock, and at seven o’clock. And the data that we have looked at showed that people did do that. But most films had a viewing window of 30 hours, which meant that people could create their own pathway in order to see loads more. So, ironically, we had people who saw more films online than they would have seen in a physical festival.
While this is certainly a relevant finding for film festivals, there is more to learn about the ways that consumption patterns shape. There are, for instance, film festivals that work with other time frames within which films could be watched, ranging from 24 hours to 72 hours or more. Data collection about audience viewing within such different time frames could be useful for festival programmers. Other relevant research could be more specifically based on the demand for collective viewing events on online platforms. That is useful to develop discussions about comparisons between the physical and online festival experience.
Advanced audience data
The forms of data analytics discussed above are relatively basic compared to what I call advanced audience data. Powerful technology-driven companies such as Netflix and Amazon rely on basic forms of data collection as well as sophisticated monitoring technology and audience profiling to acquire deep knowledge about audience behaviour and engagement with platform interfaces and catalogues.[8] They can follow several activities of audiences: from the categories through which they scroll to discover content, to the way they engage with the interface, to the content they decide to watch, to the way they consume such content, to their ratings.
Such technology-driven companies combine various forms of audience data for a variety of purposes. Some of those purposes overlap with film festivals, while others diverge from film festivals. In terms of their commercial business, one argument is that they both benefit from audience data because it can be used for marketing campaigns and advertising in order to grow their audience base. But while film festivals are at the very beginning of engaging with data about online audiences, the question is to what extent they want to capitalise on that knowledge in the long-term. As noted above by Jalladeau (Thessaloniki), they are film festivals rather than online platforms. Because film festivals can rely on basic forms of data about online viewers, there is uncertainty about the need for advanced audience data, and if such insight could generate significant commercial value. Although online and hybrid film festivals are accessible to a nation-wide audience, industry arrangements with rights holders often put limits on the number of tickets available for online audiences. In most cases, that number does not exceed the number of tickets available for physical screenings. It makes sense, therefore, to think about advanced audience data as a rather limited asset for the commercial business of film festivals.
Another argument about advanced audience data is related to content delivery strategies. The programming teams of Netflix and Amazon use advanced audience data to make production and distribution investments, as well as to provide personalised recommendations through computer-driven algorithmic technology. Film festival programmers, in contrast, rely to a greater extent on their own judgements and expertise. They make creative decisions about festival films based on cultural criteria for specific festival programmes rather than on algorithm technology and the individual audience member. Their function is associated with the traditional role of curators and gatekeepers in film and media industries. As Andrew Higson notes, they ‘function more like public service operators, presenting the curator as the guardian of “good taste”, who knows their “culture”’.[9] In line with this argument, it is often implied that decision-making of curators relies on personal expertise and professional taste. That is, however, not to say that film festival programmers should ignore audience data. It is worth exploring if, and how, audience surveys or some other form of data analytics are part of their decision-making.
Roderik Smits (University Carlos III of Madrid)
References
Alexander, N. ‘Catered to Your Future Self: Netflix’s “Predictive Personalization” and the Mathematization of Taste’ in The Netflix effect: Technology and entertainment in the 21st century, edited by K. McDonald and D. Smith-Rowsey. New York: Bloomsbury, 2016: 81-97.
Arnold, S. ‘Netflix and the Myth of Choice/Participation/Autonomy’ in The Netflix effect: Technology and entertainment in the 21st century, edited by K. McDonald and D. Smith-Rowsey. New York: Bloomsbury, 2016: 49-62.
Beer, D. ‘Envisioning the Power of Data Analytics’, Information, Communication & Society, Vol. 21, No. 3, 2018: 465-479.
Higson, A. ‘Workshop 1: A Research Report’, 12 December 2016: https:// ondemandproject.wordpress.com/workshop-1/ (accessed 18 February 2022).
Livingstone, S. ‘Audiences in an Age of Datafication: Critical Questions for Media Research’, Television and New Media, Vol. 20, No. 2, 2018: 170-183.
Rhyne, R. ‘Film Festival Circuits and Stakeholders’ in Film festival yearbook 1: The festival circuit, edited by D. Iordanova and R. Rhyne. St. Andrews: St. Andrews Film Studies, 2009: 9-39.
Smith, M.D. and Telang, R. Streaming, sharing, stealing: Big data and the future of entertainment. Cambridge: MIT Press, 2016.
Van Dijck, J. ‘Datafication, Dataism and Dataveillance: Big Data Between Scientific Paradigm and Ideology’, Surveillance & Society, Vol. 12, No. 2, 2013:197-208.
[1] Van Dijck 2013; Livingstone 2019.
[2] Alexander 2016; Arnold 2016.
[3] Rhyne 2009, p. 19.
[4] There are various ways through which film festivals can gather data about audience demographics. For instance, audience demographics can be based on information included in the registration accounts of individual audience members.
[5] Beer 2018.
[6] Interviews were conducted in Spring 2021. They were part of a research project about European film festivals in times of COVID: https://www.filmfestival.gr/attachments/article/27836/study-european-film-festivals-in-transition.pdf
[7] Some film festivals, such as Göteborg in February 2021, worked with their own in-house online platforms. Other festivals relied on the services of external platform providers to build and design online platforms. Thessaloniki worked with the external online provider Shift72 in November 2020, and Dublin with Eventive in March 2021. Both in-house platforms and external platforms can provide insight into audience data for film festivals.
[8] Smith & Telang 2016; Alexander 2016
[9] Higson 2016.
Trackbacks & Pingbacks
[…] Sources 1 – 2 – 3 – 4 […]
Comments are closed.