Streamers: Forget the paradox of choice
Evolutionary models will help us find something to watch.
EXECUTIVE SUMMARY
If you are reading this, it is likely you are an executive at a streaming service, and you already know you have a big challenge. Your large content library is paradoxically an enormous draw for your users and a big obstacle to discovery, and therefore to retention and growth. Once users get past the top 100 titles, they have great difficulty “finding something to watch.” This white paper is about solving that problem.
When my team launched MGM’s movie channel, MGM HD, we had a similar challenge: to monetize the older, lesser-known titles lying fallow in the deep library. The key was to give viewers a reason to watch content they had never heard of, and to gain their trust in our “recommendations.” To that end, I ran a series of Promax-nominated short-form pieces curating the library, and keeping viewers engaged on channel between films. This was a great success, and proof of the power of story.
Generalizing this success to include streaming platforms requires a deeper understanding of decision-making and what guides non-linear, sequential choices in human beings. Below, we’ll explore those evolved mechanisms of choice, compare them with strategies scaffolding choice on streaming platforms, and examine the gaps between the two. Finally, we’ll see how a combination of recommendations and narratives can be used to close those gaps, reduce churn, and fulfill the brand promise of deep libraries.
This is not a simple problem to solve or even to understand. If it were, it would have been figured out long ago. What follows then, is by necessity, a deep dive into complex and unfamiliar territory, but one well worth the effort.
Let’s get started.
The difficulty of “finding something to watch” on streaming platforms has become an urgent, growth-limiting drag on an otherwise burgeoning segment of the entertainment industry. In general, the problem is diagnosed as a “paradox of choice,” which states that as the number of titles in a library goes up, users’ ability to choose among those titles goes down. Even as Amazon Prime Video adds the entire MGM library to its offerings, the prevailing prescription for this paradox is to limit content options while driving choice through personalized recommendation. This strategy is good for attracting new users, but it is mismatched to the longer-term objective of monetizing deep libraries through user retention. The challenge of the next frontier in streaming then is to crack this problem of “finding something to watch,” and market leaders will need an innovative solution to do it.
To that end, what follows is a synthesis of insights gleaned mainly from the fields of Foraging Theory, Evolutionary Psychology, and Narrative Analysis, culminating in a series of recommendations to improve discovery. It is a user-centered approach derived from a simple given: our evolved psychology should guide the shape of our technological solutions, not the other way around.
Roughly speaking, Evolutionary Psychology extends the familiar concept of adaptation to the province of the mind, identifying evolved neurological structures and processes that drive human behavior, including choice. A related discipline, Foraging Theory, is a subfield of Behavioral Ecology, concerned with how and why animals choose food items, and the strategies they use to balance the expenditure of time and energy to find them. Narrative analysis in this context is simply the process of recasting “stories” as the medium in which human beings encode, store, and transmit the consequences of choices to drive future decision-making.
IT’S THE STORY
Let’s begin with the conclusion: the largest gap between our strategies on-platform and our users is the lack of meaningful and efficient narratives to guide choice. Narrative solutions may take many forms but those specifics are secondary.
If narrative seems like an odd point of entry to the process of choice, consider why stories exist. They are an end product of the process of choice and, because they are instructive in nature, they also drive it. They are post mortem reports designed to guide future choices. On streaming platforms, we might also think of narratives as the “why to watch” currently missing from the “what to watch” of recommendations.
Readers of Carl Jung, Joseph Campbell, or even Robert McKee will already know that a story has a prescribed shape with the following key elements: 1) A problem arises in the form of the unknown, 2) A hero explores the problem and finds a solution, 3) The hero wins a treasure in the form of improved knowledge about how to live in the world, 4) The hero shares this knowledge with the wider community.
Now, let’s compare those story elements to the process of human choice:
1) A problem arises in the form of an unknown option, 2) Exploration of the unknown option creates subjective value judgments, 3) Ranking the value of the unknown option against a hierarchy of known options reveals a clear choice, 4) The consequences of the choice are used to update the value of the chosen option, 5) The sequence above is encoded as a narrative to be referred to during future decisions or shared with others.
The takeaway from this comparison is that narratives take the shape of choice-making because they are molded by it and also because the informational components they contain need to fit back into the process of choice in the future. Like a map with a route marked in red, a narrative’s function is to guide us through forking paths of decision toward the goal of minimal harm and maximum benefit.
WHAT’S THE PROBLEM?
Before we get into retooling streaming platforms, let me invoke an old chestnut technical-types like to use on bell-and-whistle-loving creatives. It goes like this: “Technology is the solution to a problem. So, what’s the problem?” Despite having been on the wrong end of this barb myself, I love its demand for congruity between our understanding of a problem and any potential solution before it is proposed. So, in that vein, let’s make sure we understand the problem of “finding something to watch,” and where it came from.
In the early 2000s, the job of a content management system (CMS) was to take a personal library of known, familiar content off groaning bookshelves and put it on a personal computer. Online purchases were soon added, but in a nutshell, the problem the CMS was designed to solve was, “How to store and organize familiar content.” Programs like iTunes worked well to curate such libraries, and soon became a standard for similar content management systems. Then, a subtle change occurred.
In 2007, Netflix offered its subscribers access to 1000 streaming titles. Its CMS looked similar to iTunes but an important change took place beneath the surface. The unfamiliar nature of this content silently switched the problem to be solved from how to store and organize familiar content to how to find something to watch among unfamiliar content. It was this switch to unfamiliar content that opened up a mysterious “lack” in our ability to choose.
Recommendation was put forward as the solution — narrowing options on a personalized basis, and thereby shifting the heavy lifting of choice to the platform side of the user-platform interaction. A simplified version of algorithmic choice looks like this:
While this process mirrors the process of human narrative and choice outlined previously, data shows that once users receive recommended options, the suite of curation, editorial, marketing materials and metadata simply doesn’t drive choice among those titles very well. There are good reasons for this, which will be explored further, but the result is that the goal of “finding something to watch” remains yoked by many of the same challenges it had coming out of the gate. At a streamer with a library of 35,000 titles, just 100 of them may account for 50% of views. In the comedy genre, the top 10 licensed comedies account for 80% of views. And once that low-hanging fruit has been picked, users are left to forage on their own, or accept recommendations, which only 26% of them do. Netflix claims 80% of stream time comes from recommendation, but if we turn the framing effect around, that means, on a platform where 100% of visible options are recommended, 20% of stream time comes from navigating around those recommendations. Search times across platforms still average roughly 9 minutes, search failure rates hover stubbornly at 20%, and recent churn rates (i.e. users quitting their subscriptions) range between 13–20%. These numbers point to a persistent disconnect between users and content on platforms — a roadblock to the retention needed to drive superior revenue growth.
One of the principal reasons for this problem lies in the missing link between “options” and “choices.” These words are often used interchangeably, but they are not the same. If we think of a restaurant, the menu is a list of “options,” while what we order from that menu is a “choice.” In the middle sits a diner hopelessly trying to decide between unfamiliar dishes based on scant descriptions. To extend the metaphor, a waiter offers to help by recommending the seabass, but our dubious diner says, “Give me 9 minutes,” which she then spends rescanning the options, sweating the FOMO of ordering a predictable dish she’s tired of before finally succumbing and ordering it anyway.
Is this not how, on platforms bursting with content, we end up binging The Office for a third time? Let’s examine why that happens and look for the gaps between user and platform that might indicate where a fix is needed.
It’s true, we are missing a “narrative bridge” between options and choice, but we should backtrack for a moment. What if our waiter was actually right and, had she listened, our diner really would have loved that seabass? This highlights two points. One, that accurate recommendation of titles is not the same thing as giving the user a compelling reason to choose those titles, and two, that narrative does not need to replace recommendation, it needs to complement it. To give it a voice.
What we need then is a knowledgeable, confidence-inspiring waiter with a great patter and a sizzling skillet of fajitas he can hold under our noses, prompting us to say, “Ooh. That looks good.”
But before we can determine “the how” of executing such a thing, we need to learn more about “the why” that will ultimately shape it. We’ll find the information we need down among the foundation stones of choice.
TURNING OPTIONS INTO CHOICES
If a tree falls in the woods and no one is there to hear it, does it make a sound?
Choice begins in the intertidal world where objective reality meets subjective value. Where the existence of something is strangely difficult to disentangle from our observation of it and its utility to us.
Like a beachcomber inspecting curiosities that have washed ashore, we turn objects over in our minds until a felt sense emerges that the thing is either charged with positive, neutral, or negative emotional value, based on what it can do to us or for us. In the case of neutrality, the object becomes irrelevant. Otherwise, this charge or “valence” imbues the object with a psychologically repellant push or attractive pull, the strength of which is determined by the intensity of our initial reaction, and by the object’s usefulness in achieving goals in a given context. In this way, we turn an emotional response to an unfamiliar object into a quantifiable data point that can be ranked among many others, on a continuum from “bad” to “good”. The accumulation of these ranked data points forms a kind of database of options called a hierarchy of value.
This hierarchy’s job is to cross-reference a goal, the current context, and the available options, with the aim of elevating one option above the rest, so that it becomes visible as a clear choice. That is, the option which minimizes potential harm and maximizes potential benefit. Or in the case of streaming platforms, the title which maximizes our satisfaction.
WHAT IS THIS TO ME?
As we look out on the world, the first filter that our brains apply is the binary of explored/unexplored, wherein the explored is known, familiar and positive, and the unexplored is unknown, unfamiliar, and negative. In the same way that a stranger we are wary of can become a great friend, the transformational process of exploration is about turning the unfamiliar into the familiar, and the negative into the positive (or at least “the known”).
This is the way objects take on values, but our goal as humans is not just to admire the intrinsic values of things, it is to understand how they can help us get what we want (i.e. “goal-directed behavior”). We do this by asking one simple question about an object, “What is this to me?”
What an object, “is to me” is not about what it looks or feels like. It is about its potential uses. For example, the answer to, “What is a stone to me?” is not a description like, “hard, heavy, round object”. Rather, it is a description like, “It can smash open walnuts,” which connects to a goal, like eating, and a benefit, like nourishment.
For titles on a streaming platform the question, “What is this to me?” also connects us to potential goals and benefits. For example, in the context of a date night where the goal is to “Netflix and chill,” we need a thing that “sets a romantic mood.” When — among the rows of key art — we see a title like The Notebook, it rises to the top of our hierarchy of values for this context and we say, “Yes. This is the best thing that sets a romantic mood.” A choice is then made that connects us to a desired outcome.
But what about the world of unfamiliar content? How do we solve our problem of “finding something to watch,” when we have no idea what any title, “is to me,” and the platform lacks a mechanism for meaningful exploration to find out?
THE THREE-WAY TUG-O-WAR (CURIOSITY, HOPE, & ANXIETY)
Impressive as it is, the human brain is not powerful enough to engage directly with the firehose of reality coming at us through our five senses. Instead, we simply check our perceptions against mental models of what the “normal world” should look like. Novelties that don’t match up are uncategorizable as either dangerous, beneficial, or irrelevant, meaning we have no idea “what they are to us.” Just as in a narrative, the emergence of such an unknown “problem” is a call to action and exploration, arousing a cocktail of emotions — namely curiosity, hope, and anxiety.
To understand the three-way tug-o-war this inflicts on our minds, just imagine what goes through a dog’s head when a scary-looking stranger dangles a treat in front of its nose. What is that thing in his hand? Food? A rock? Some kind of trick? The animal approaches hesitantly, alternately sniffing at the thing and retreating; curious to know what is being offered, hopeful that it is food, and anxious that the stranger might pull some dangerous ruse. To a lesser extent, these are the same forces aroused in us by the emergence of unfamiliar titles. Like the dog we have a desire to explore such objects until they are either rejected as irrelevant or risky or they become known, positive, and worthy of consuming.
If we compare this mental process to the strategy scaffolding choice on streaming platforms, a gap becomes obvious. Without a way to truly “sniff” the titles dangled before us by the “stranger of recommendation,” our hope and curiosity have nothing to entice them. There’s just an anxiety-inducing hand waving unscented key art under our noses. We scroll across row-on-row of valueless options, a swirling M.C. Escher staircase to nowhere. Hope and curiosity wither. That paralyzing “paradox of choice” rears its ugly head, and “click” — we retreat from the platform like a skittish dog.
While such anxiety may bias us against the unknown, there is a countervailing exploratory urge. Call it “hunger.” It’s a powerful motivator. We just need a way to direct our natural appetite for content.
THE PROBLEM OF WHERE TO GO NEXT
The name of the game in Foraging Theory is optimization. Typically, this is applied to gathering food, but for our purposes, optimization means minimizing the time spent in discovery, while maximizing the time spent viewing content. But not just any content. Good content.
While a foraging animal is balancing energy expended versus energy consumed, humans hunting for content are balancing a currency more like “satisfaction” — burning it in the hunt and restoring it through viewing.
We might say there are two success criteria in the foraging process, quality and quantity, and decisions based on these criteria operate in a sequential and nested fashion. Let’s imagine that a user stumbles upon seven seasons of Game of Thrones on HBO Max. The first choice the user is faced with is called the “accept-reject problem.” Simply put it means, “Is this content worthy of my time and energy?” Presuming the answer is yes, the second choice is the “stay-switch problem,” which means, “How long should I continue watching?” Which is another way of saying, “Is there a better scene, episode, season, series, curatorial category, or platform to watch?” Stay-switch is the roving eye that keeps us forever wondering if the grass is greener on the other side, which is a good disposition for constantly optimizing our choices.
All of these stay-switch dynamics are captured in the elegantly simple Marginal Value Theorem (MVT), which tells us that: a user should switch platforms, curatorial categories, etc., when the reward rate drops below the average reward rate of the environment (for additional detail on MVT, see Exhibit 2, below).
MVT is useful because it illuminates the way rising discovery costs and diminishing viewing benefits set a threshold for stay-switch decisions. Its weakness however, is that it ignores how our thinking about, “where to go next,” also moves that stay-switch threshold. In other words, if we know a better spot to forage, we’ll quit the one we’re in sooner. Keeping a mental list of such places that would be good “to go next” creates a curated archipelago of foraging areas of above-average value. Charting a path between these spaces then lifts the average value of the environment.
Is this not what streamers are trying to do for users? Whittle a large, uneven library down to curated sets of the best things to watch for a given user? Only, with streamers, the whittling is primarily happening on the platform side of the user-platform interaction (recommendation). This cuts off the predictive power of the user by hiding the values of options inside the black box of an algorithm, rather than placing them in narratives which can feed back into the decision process.
The last factor to mention in this process is “prey handling,” — a measure of how difficult it is to process food once it is gathered. For example, coconuts require a lot of work to open, and there’s always a chance that what’s inside is rotten. So finding a pile of them doesn’t mean we’re eating anytime soon. Similarly, users on streaming platforms have a large pile of content at their fingertips but exploring it requires a lot of “cracking and peeling” to get at the satisfaction of watching good titles. This drops rates of consumption, satisfaction, and retention. We need ways to speed up this handling or exploration.
It would be great if we could simply hand users a hierarchy of value in the form of recommendations, but as it turns out we humans require the story of how a hierarchy was formed in order for it to hold any meaning (unless the recommendations come from a very trusted source, in which case we need a story about the source). A set of options with no organizing story is something like a body without a skeleton to arrange and animate it. Yet, this is the current condition of recommendation on platforms. All “what” and no “why.” No narrative to help us understand, “what is this to me?” or if the outcomes of watching might be a fit for our goals.
THE DRIVER OF THE PROCESS
You might have noticed above that the more great content we have to pick from, the choosier we become (that stay-switch threshold moves lower). This sounds like a good problem to have, but it is a precarious moment for users — a fork in the road. Down one path lies choice paralysis. Down the other lies narrative, with its ability to create meaningful arrangements of titles and lift hierarchies of value.
Lowering the stay-switch threshold with an abundance of great content then creates a commensurate obligation to help users understand why certain titles are great. This is what we really mean when we say, “find something to watch.”
Like a Bronte heroine trailed by adoring suitors, the fickle user of various streaming platforms will forever be moving through cycles of “accept-reject” and “stay-switch.” So, platforms need to be constantly on the woo, winning the user with stories, raising hierarchies of value, and all-the-while making a deeper connection by revealing a new side to the platform — a trusted persona who is wise, insightful and always there.
THE SHAPE OF THE SOLUTION
If you’ve ever strolled through an art museum you’ve probably had the, “What am I looking at here?” feeling that comes with wandering aimlessly between unfamiliar works, wishing you knew more about them. If you are lucky, you cross paths with a group led by a docent spinning yarns about, say, Modern Art’s break from tradition, and the heated rivalry between Picasso and Matisse — each of their paintings landing like a punch and counterpunch in a boxing match that built a new genre. She ends on Picasso’s darkly humorous eulogy for his urinal-as-art nemesis, Duchamp and the group laughs. Suddenly, the paintings in front of you that were dry and meaningless are swimming in the sauce of story. And as the group moves on, what do you do? You follow, eager for more.
Right now, curation on streaming platforms is less like this docent and more like the sign hanging over the entrance to the gallery that reads, “Modern Art.” That sign might get us into a room full of thoughtfully arranged paintings, but we still need the docent to help give them meaning. On platforms, we need an authoritative, trusted voice to speak to “why to watch,” build a connection with users, reinforce recommendation, and help us build those hierarchies of value that drive choices. This is how we will exploit the long tail of the deep library, reduce churn, and capture the lifetime value of the user.
Getting too specific about what those narratives look like in the absence of testing data is an exercise in guess work, but taking a cue from foraging theory, we can predict which patches are likely to be fruitful and worthy of our time — what I’m calling the “shape of the solution.” It’s important to remember that the “what” of a video — its content — should not be looked at in isolation. Equally important are “where” it is placed on the presentation layer, “when,” and “why.” We need to examine these four ‘W’s in service of the fifth and most important ‘W’, the “who” — our user. Any narrative content we serve up should be thought of as a response to her exploratory needs in the moment as she forages for something to watch.
The below interstitial from my days at MGM HD, Craft: Marlon Brando, might serve as a rough approximation of such a narrative piece.
Let’s say that our user is a woman who’s been on a film noir jag. She’s recently watched or searched for Chinatown, The Maltese Falcon, and The Third Man but tonight, she’s reached the end of the Film Noir category without making a choice. So, what now? Should she switch categories? Switch platforms? Scroll aimlessly?
Here’s an idea: what if at the end of that row, she discovers a thumbnail for a piece of curatorial content on Neo Noir and it begins to autoplay? The burden of figuring out “where to go next” is eased by offering a helpful mode of passive and entertaining exploration. Perhaps the expert “docent” of this experience sketches out the conventions of Film Noir; how films and series like Altered Carbon, Croupier, Night Crawler, Animal Kingdom, and Brick fit in and what makes them unique, along with a little color about how these stories and performances were conceived and produced. Just enough to whet the appetite while remaining light on its feet.
As each film is introduced, a simple UI element appears in the lower-third with options to “watch now,” “add to list,” or “notify when available.” We might even have a “subscribe” button to turn free AVOD users into paid subscribers. After 2–3 minutes, the piece ends with a rousing outro, and a call to action to watch a title, go back, or to pick another curatorial piece based on user history. For example, if our user-added Night Crawler to her list, a thumbnail for a journey through Jake Gyllenhaal’s films or films with anti-heroes might appear. Instead of going back to the home screen, she might do a little meta-browsing through narrative pieces, right there.
Now, let’s imagine another user searching vertically through the rows and descending into less and less relevant options. Instead of letting that continue, why not allow him to hover over or click on a curatorial category like “Coming of Age Stories,” and trigger a piece of modular, curatorial content tailored to his history? He’s already seen some of the titles in the piece — Lady Bird and Superbad, so, per recommendation algorithms, those are swapped out for Book Smart and The Way, Way Back, which fit right in with the remaining titles, Badlands, Harold and Maude, and Brighton Beach Memoirs. After a funny and touching 2–3 minute look at what makes these stories tick, and how they broke new ground, our user receives the same call to action as above and can even hover over the featured titles to bring up pieces of curatorial content with intersecting themes. For example, hovering over Book Smart might bring up a piece on buddy movies, or one on comedies with a feminist slant (for a breakdown of additional features and strategy see Exhibit 1 below).
In each case, users are served up a highly relevant list of titles, wrapped in a story that frames them in an insightful way and tells us why they are worth watching. This frictionless delivery gives users the raw material to build their own hierarchy of values, and make choices based on our curation, combining the power of algorithmic and human choice.
THE FUTURE
The next great frontier in streaming will belong to services that can solve the problem of “finding something to watch.” They will tap the potential of narrative to arrange options in hierarchies of value, giving users the “why to watch” currently missing from recommendation’s “what to watch,” and thereby connecting users to satisfying choices. By attending to the foraging problems of “accept-reject” and “stay-switch” through story, innovative services will be able to keep the user’s roving eye from wandering off platform or becoming dissatisfied. And by creating a trusted docent and voice of the platform, they will drive exploration and choice, deepen the connection to the brand, increase retention and, ultimately, give greater meaning to the user experience.
As someone deeply invested in the intersection of storytelling, technology, culture and cognition, I am excited to leverage narrative to improve that experience; and, like everyone else, I look forward to finding something great to watch.
About the author: Jeremy Cole is a marketing strategist and content expert with a passion for culture, evolutionary psychology, and storytelling. He holds a BS in anthropology. As Executive Director of Creative Services for MGM and Creative Lead for the movie channel, MGM HD, he has created Promax nominated interstitial content and been immersed in thinking about engagement and the curation of deep and historic libraries for over a decade. His varied career also includes stints as an assistant to Steven Spielberg, a story editor, a documentary writer-producer, and a creative director for location-based entertainment projects. He has been instrumental in launching innovative smartphone tours, interactive exhibits, and a VOD sales platform. He can be reached at jeremycole@me.com and https://www.linkedin.com/in/jeremymcole/
EXHIBITS
Exhibit 1: KEY FEATURES OF NARRATIVE PIECES
The narratives discussed in the section “The Shape of the Solution,” above should take the form of stories with the following qualities:
VIDEO-BASED
Video is a highly engaging, passive, and efficient means of exploration with very little cost for the user. If done well, such content is not only relatively frictionless, but additive and enjoyable. These videos must be:
- Short. The object is to whet the appetite, not satisfy it
- Efficient. Multi-title pieces create large, novel groupings of content around surprising, insightful and delightful, stunts, themes or frames
- Cascading. Themes intersect, diverge and fork, leading users on endless contextual journeys
- Persistent. Driving choice now and across time with opportunities to watch now, add titles to lists, share, and opt-in to reminders and notifications about availability/windows/new seasons
- Complimentary. Give the “why to watch” to recommendations “what to watch”
PERSONALIZATION
Narrative pieces are relevant evergreen videos with optional, swappable, modular segments, allowing them to be refreshed and allowing for granular personalization. These pieces must be:
- Data-informed. Leveraging historical user data
- Synergetic. Creating a data-rich feedback loop to improve recommendation, engagement, and trust through interaction with the user
- Interactive. Offering simple UI elements and calls to action to mark titles for watching now, adding to lists, sharing, alerting when available, etc.
PERSONA
Humans watch the consequences of other people’s choices (their stories) in order to decide if they are worthy of imitation (trusted). It is impossible to observe this same thing in recommendation algorithms, and therefore difficult to trust them, but with narratives we can give recommendation a human face. To do this we must:
- Create a persona for users to project the quality of wisdom/trust upon — a storyteller.
- Demonstrate deep insight, expertise, and positive outcomes
- Surface the factors behind algorithmic choice through story — give a “why” to the “what”
- Deepen the user’s connection to the brand in the mode of backend content marketing
- Fulfill the brand promise of exciting new originals plus a deep library
STRUCTURE
- Give a spiral structure to narratives.
- Provide users with the same information first in abbreviated form, then in longer form. This allows choosing quickly or diving deeper.
- Calls to action and other features spiral with the narrative, providing off-ramps to choice.
DISCOVERY
Pieces must be thoughtfully integrated into product design and the user journey. Possibilities to test include:
- Factor user time on platform or a 40 title browsing limit to trigger autoplay of a piece or a prompt to watch
- Autoplay at the end of an unsuccessful search in a curatorial category
- Trigger a genre piece by hovering over or selecting a curatorial category
- Leverage fast/emotional choice to break decision difficulty and increase satisfaction. When user’s cognitive load is high, introduce time pressure — a limited time offer to watch
- Interrupt the bias. After quitting a title we dislike, we are likely to choose a “same old” sure bet. Interrupt this by triggering a narrative piece, offering relevant, passive exploration
Exhibit 2: MARGINAL VALUE THEOREM
A model of the cost/benefit curve in animal foraging that has been adapted to content discovery on streaming platforms. It can be used to predict when and why users will switch platforms or curatorial categories. As an equation, it is expressed as:
W = B(C)/C
Where the benefit of the cost, B(C), represents net satisfaction gain as a function of cumulative time spent foraging on a platform, W represents that net benefit divided by foraging time plus transit time between platforms or curatorial categories.
The benefit curve above depicts the diminishing returns in satisfaction of consuming content on a single platform or curatorial category over time. The intersection of the diagonal line and dashed vertical line marks the optimum point in this benefit curve for a user to quit the current location in favor of another. User satisfaction then is determined by the steepness of the diagonal line. (Image credit: David W. Stephens)
___________________________________________________________________
ACKNOWLEDGEMENTS
Alvino, Chris and Basilico, Justin: Learning a Personalized Homepage. Netflix Technology Blog, 2015.
Hall-McMaster, S. and Luyckx, F.: Revisiting Foraging Approaches in Neuroscience. Department of Experimental Psychology, Oxford University, 2019.
Neumann, Erich: The Origins and History of Consciousness. Princeton University Press, 1973.
Parks and Associate: Churn rate drops to 38% among OTT services and 49% among vMVPDs, 2020. https://www.parksassociates.com/blog/article/pr-12082020
Peterson, Jordan B.: Maps of Meaning: The Architecture of Belief. New York ; London : Routledge,1999.
Schwartz, Barry: The Paradox of Choice: Why More Is Less. New York :Ecco, 2004.
Sinek, Simon. Start with Why. Penguin Books, 2011.
Stephens, David W.: Decision Ecology: Foraging and the Ecology of Animal Decision Making. Cognitive, Affective, & Behavioral Neuroscience. University of Minnesota, Saint Paul, Minnesota, 2008.
Tversky, Amos and Kahneman, Daniel: The Framing of Decisions and the Psychology of Choice, 1981.
Warner Media: HBO Max Investor Presentation, October 2019
