week 7 Flashcards
Video Streaming Services
(also known as Video-on-Demand services) are digital
platforms that provide users with on-demand access to a wide range of
predominantly series and movies over the internet, usually through a subscription-
based model
Netflix Content Acquisition Strategy
Then Netflix developed the binge model:
Convenience: watch at their own pace, binging the content if preferred Immediate gratification: there’s no waiting needed
* Netflix almost always pays a fixed fee for regional rights to stream a movie or
series for a limited time.
* Netflix applies insights from data analytics to select movies and series that they
want to add to their library.
* Data analytics can describe the conditions under which certain decisions have
been successful. It thereby provides a perspective that supports decisions with a
higher probability that these products will be successful.
Weekly Rollout
- Sustained engagement
Sense of anticipation
Extended engagement - Content Promotion
Prolong conversations
Episode-specific reviews - Watercooler Effect
Stimulates social interactions
Shared viewing experience
Binge Model
- Convenience
Watch at their own pace
Binging if preferred - Immediate gratification
No waiting
Enhanced viewing
experience and satisfaction.
Familiar things tend to be more enjoyable, new things tend to be more
interesting.
Exposure to familiar media content provided affectively similar experiences to
pleasurable entertainment in terms of enjoyment, and in terms of
meaningfulness.
Algorithmic Recommendation
Familiar things tend to be more enjoyable- that’s why many of the products have sequels. -Collaborative recommendation systems are algorithmic tools that are used to identify and recommend content that may be of interest to users: there’s an information overload (thus we also use personalized recommendations).
Personalized recommendations are considered the most effective way to
influence the selection process and solve the problem of information overload
(Porcel & Herrera-Viedma, 2010).
About 20% of the hours spent on Netflix result from searches on the site, while
about 80% is inspired by popularity rankings and recommendations
(Gomez-Uribe & Hunt, 2015).
Algorithmic Recommendation 3 types
- User-to-user recommendation finds other users whose preference is similar to a target
user and then predicts to what degree this user will like a product that he or she has not
yet experienced, based on the choices of those “taste neighbors” - Item-to-item recommendation uses similarities between rating patterns of items rather
than between individual users. Although this connection is also based on input from other
users, the algorithm focuses on the ‘item neighbors’, not the taste of similar individuals. - Content-based recommendations use relevant inherent attributes of a product,
not other consumers’ input, as the source of similarity. Products whose relevant
attributes match those of products liked by a consumer are recommended, while those with different attributes are not.
Cold Start Problem
Sensitivity theory is similar to uses and gratifications theory: but adds that not only we consume certain products depending on our need but also to a certain extent.
We always look for new things, other wise it get boring and the algorithm shows us the same things
- All recommendation systems face the “cold-start” problem: how to deal with new
products (for which no consumer ratings yet exist) or new users of the system
(who have not rated/watched any products themselves)? - Content-based recommendations can provide recommendations even after
one single selection (but require elaborate coding beforehand) . - An advantage of item-to-item over user-to-user recommendations is that it
requires less calculation, time and effort. Recommendations can cascade from
the first selection onwards. User-to-user recommendations require more
elaborate selections from a user to develop a sense of group sorting (groups are
too big at first to be meaningful)
Monotonous Recommendation Problem
- Algorithmic feedback loops generally limit the ability to explore films from a
variety of genres or themes. Algorithms keep viewers confined to their
(cultural) bubble, for instance by recommending American films to American
audiences (Jin, 2021). - User-to-user recommendations have a higher potential for serendipity
(compared to Item-to-Item or content-based recommendations) because any
surprise discovery made by a neighboring group member has a chance to
spread to other users of that group via the algorithm. - However, most filtering algorithms effectively provide preferred yet diverse
content (Araujo et al., 2020) - Familiar stimuli can become predictable and boring, whereas novel stimuli that
are not fully understood elicit interest, which engages the action tendency of
exploring new and exciting experiences (Tan, 2008). - Sensitivity theory (Reiss & Wiltz, 2004) states that individuals differ in both the
types of reinforcement they desire and in the amounts of reinforcement they
need to satiate those desires. - Users pay attention to stimuli that are relevant to their motives, and these
motives tend to be intertwined with underlying personality traits and moods.
How we choose on Netflix
- Anchoring is a cognitive bias that describes the common human tendency to
rely heavily on the first piece of information offered when making decisions. - The serial position effect states that when given a list of information and later
asked to recall that information, the items at the beginning (primacy) and the
items at the end (recency) are most likely to be recalled than items in the middl
User-generated Eudemonic videos
They wanna watch meaningful content, not just hedonic
Limited viewing options: shadowbanning refers to restricting the access to content without their knowledge. TikTok and Insta ban multiple products even about the war.
Live streamers are increasingly more popular: they respond and engage with viewers in the chats. Content analysis showed a lot of sexualisation in livestreams.
-Gen Z and the digital native generation: we wanna be connected the whole time
DIgital liveness generates a sense of unpredictable flow and potential eventfulness during cultural events (you wanna be part of it).
Liveness:
Binge watching
Binge watching is not clearly defined. Some define a binge as watching 2–3 episodes without a break- most people that watch series binge watch.
-88% of Netflix subscribers watch at least three episodes of the same series in one day
-it’s often unintentional
Revenue for streaming platforms comes from subscriptions. -Netflix is struggling.
What motivates people to post content?
You wanna be unique and how to be unique: you can use memes referring to self replicating communication in digital media that convey cultural ideas or contents and often change and vary.
- We desire to express our identity in a social affirmation People react to people
What makes a video go viral?
-Viral video is defined as a clip that spreads to the masses like an epidemic through social media.
Sharing has social currency: means that if people find the video we posted interesting, we become interesting
Distinctiveness determines if a video becomes viral or not: has to be distinctive and standing out. Also Novelty plays an important role, regardless of the actual content.
In general content that evokes anger or amusement is more likely shared: content that’s arousing regardless of why it is arousing, get shared.
Rage-bait refers to comytent that deliberately elicits outrage with thee goal of increasing internet traffic Lies spread faster than truth: often because you just add fake information to make it more interesting.
The Future of entertainment
Web 3.0 is the third generation internet
