netflix content reviews

netflix content reviews

Analyzing the Impact of Netflix Content Reviews: A Scholarly Perspective

1. Introduction to Netflix and Content Reviews

Over the past several years, Netflix has been actively researching and improving its content recommendation algorithms. For several years, Netflix has run a competition called the Netflix Prize in which teams competed to design the most accurate recommender algorithm. In 2012, Netflix announced it would now crowdfund its content-recommender algorithm, with a prize pool of one million. As a result, Netflix is now able to predict content (movies) quality. Netflix possesses house-generated critic ratings of content (from Variety Magazine) and user-generated ratings from the consumer market, which allow comparing the coverage, intensity, and effect of both critics and users within the American VoD market.

Netflix, Inc. is a manufacturer and provider of services that stream television shows and films to its customers via the internet. It began as a United States-only DVD-by-mail service in 1998 and started streaming content in 2007. Since that time, Netflix has expanded beyond the United States and now offers service in 41 countries. As of the end of the year 2011, Netflix had over 22 million customers who watched over two billion hours of television and films per quarter. To expand its worldwide presence and viewership, Netflix recently announced its first global licensing agreement with Nu Image to own all first-run rights exclusively to hundreds of movies internationally.

2. The Importance of Content Reviews in the Streaming Industry

The majority of productions have received Netflix’s internal age rating and content descriptors with the assistance of Netflix’s high-speed, high-quality, and high-scale content rating operation. This operation leverages a unique combination of trained adult taggers who understand a variety of international cultural nuances, trained young-adult raters who reflect the qualities and interests of the depicted on-screen content, and an efficient automation infrastructure that accelerates many rating operations. During the rating process, taggers and raters provide credibility, consensus, and expertise that help ensure that the resulting data closely reflect the sensitivities, knowledge, and concerns of Netflix’s global audience and the teen and adult viewers who make most Netflix streaming decisions in each global market.

Netflix and other companies within the global multi-channel broadcaster industry have made hundreds of original productions available to its streamers, either directly as owned-and-distributed or licensed content. Dozens of new titles are offered every month in the U.S., and hundreds of non-English debuting titles occur each year in international markets. These titles vary widely in storylines and suitability for streamers of all ages. Given the need of parents and guardians to make informed viewing decisions for children and of adult viewers to make informed viewing decisions for themselves, every title comes with a set of detailed content ratings.

3. Methodologies for Analyzing and Evaluating Netflix Content Reviews

3.1 Content Analysis Content analysis is the systematic and replicable examination of any content such as books, paintings, journals, newspapers, documents, and transcripts. There are some problems with content analysis. One of these is that, as before in marketing, when coding content the decisions about what is being looked at, as well as the context within which it is being studied, tend to be qualitative and untold. Hence criteria used to evaluate content must be defined by human judgment, a method predisposed to interpretative inconsistency. Moreover, the procedure presupposes the reliability and validity of the measures obtained from the data so that inferences, generalizations, or predictions can be made with some degree of certainty. The vast scope for human judgment indicates that measures of content do not generate reliable or valid data, although a number of methodological guidelines have been suggested that aim to ensure the quality of studies using content analysis.

There are many ways to conduct such a study. Our group has personal experience with four methods. These are content analysis, blinded content analysis, Linguistic Inquiry and Word Count (LIWC), and controlled experiments. Our current experiment in the field of negative information consumption can be added as the fifth method. We can provide insights using not only general findings, but also specific conclusions for each of these methods.

4. Key Findings and Trends in Netflix Content Reviews

While Netflix could publish a slew of numbers as Key Findings about written reviews, the shared aspects of content review writing are that reviews give members something to say, to connect, and to feel, even if it is something as simple as saying, “Have you made your mind up? Or do I need to keep talking?” (Ozark). To understand members more deeply and provide insights into new methodologies and emerging trends in the field, it is imperative to acknowledge and discuss these features in more detail. Different considerations when comparing times in the number of submitted written reviews are important along the way, such as the percent increase seen on holidays. These patterns and interesting trends in Netflix content reviews increase our understanding about not just the quantity of written reviews, but quality as well, while also adding more color to the streaming content review landscape.

Within the short 6 years Netflix has been providing members the feature of submitting reviews of Netflix content, some interesting patterns and trends have surfaced. Members are mostly writing reviews for streaming content with the quantity of written reviews accelerating farther as the amount of streaming content increases. An analysis of subscribers’ methods of writing shows that members are careful to only write about content available to watch, not spoiling others’ viewing experience. To showcase some of the popular reviews and the variety of Netflix content members are writing about, a few examples include those authored over a period of 6 years, as well as content with increased popularity and mass appeal: Queer Eye, Ozark, and Making a Murderer 2. Connecting reviews with periodicity, heated debates and cultural moments like the more recent release of Beyoncé’s Homecoming is included, along with a focus on the negative reviews and backlash from the movie Cuties.

5. Implications and Future Directions for Research in this Area

Other members of the research community might also think about responding to some of these questions, perhaps even utilizing new data sources, Netflix or otherwise. For example, how well does the typical user rating correlate with the Tomatometer average or the Metacritic score? Are differences in these “professional” impressions relatively low variance? Also, with language continually evolving and given the rich data environment the databases present, can sentiment analysis find more useful content identifiers in reviews and time-use analyses? And how can the richness of the data be used to depict the strength and range of identity in one’s own company? How large must the denominator be before a comment about your group takes on one meaning rather than another?

Viewers may wish to know which shows depict a team that communicates effectively and which shows depict communicators who are depicted as jerks or who manipulate others. They may also want to know which shows show people working together in a diverse and inclusive team. These themes could be useful in leadership and college to career courses, but there is no way to explore for content in these areas using review services and the show descriptions alone make it hard for those seeking this content to sort through episodes unless some crafty tagging has been applied.