Our Apps, Ourselves

Giving software humanity is a strategic advantage. So who will be the “human-first” tech company?

Humans are weird.

We act as individuals but move as groups. We behave irrationally, act rashly, think creatively, and love unconditionally. We’re hard-wired with a bias for narratives and justice. We have the capacity for tremendous empathy but also great evil.

Software is weird.

It does as it’s told, without much backtalk and without care for intent. It has no bias towards understanding emotion, unless it is strictly told how and when to consider it. It can do much we cannot, and yet nothing without us.

Humans and software together are really, really weird.

We act differently, communicate differently, value different things. We really just don’t get along naturally.

Nowhere is this more evident than social apps, software that is explicitly designed to extend our relationships, to give a voice to the unheard, to amplify our feelings and express them more completely or creatively. And yet, many social apps solve for human needs that do not exist. Many times, instead of enhancing our humanity, they eliminate it. Not in a Terminator sense, of course, but by confusing the user or by removing human qualities of empathy and creativity from the interaction.

The world of software is changing. The technical hurdle for creating good social products is dropping. Vast networks of scale are no longer a strong barrier to entry for communication products (see: Slack). More than ever, good design is not just about creating the cleanest-looking product — it’s about creating the product that is the most natural to use.

That’s why the best-designed, most human-oriented products now have the potential to win over the current mass-adopted solutions, especially as new platforms emerge and teenagers grow up. I don’t really care which social apps have been successful in the past or how they succeeded. Those first-movers have already moved.

Here’s what I want to know: What product is going to deliberately differentiate on humanity? Who is the “human-first” social network? Who will be the “human-first” tech company?

Endowment: Giving Our Software Humanity

One way we’ve bridged the gap between software and human is by endowing our software with personality. People relate innately to other people, especially if they see themselves reflected. If we sense human character in the software we use, it makes our experiences feel natural and we react positively. At our core, we are social beings.

We’re pretty easy to fool, especially with simple interfaces like text. If you talk with a well-designed chatbot, the experience is often pleasant and effective. Slackbot, Slack’s friendly onboarding bot, is one example of chatbot design done right.

But Slack has also created a vast platform for chat-based applications to thrive, and the response has been inspiring. With the emergence of tools like Botkit, creating powerful interactions no longer requires software designers to have a strong sense of visual design or complex code so much as it requires a deep understanding of human communication and a large dose of empathy for the consumer. Bots don’t have to do and understand everything — they just have to be able to serve one problem or need particularly well. Asking our bots to do too much actually compromises their illusion of humanity as well as their usefulness.

With this emergence of chat platforms and tools to support development, creating powerful experiences, utilities, and businesses is no longer solely the domain of elite technical workers. Anyone with emotional intelligence and an understanding of the humans behind the needs — and particularly how these humans talk about their needs — can build incredible products.

This is what I’m calling “communication-market fit,” not “product-market fit.”

The danger to this is what I’ve written about before: the “uncanny valley” of social communication. If we attempt to create human-like chat experiences with software algorithmically, the software must not try to convince the user that it is human unless it can mimic human interaction perfectly.Surprise: nothing can… yet.

Slack calls the chat-based applications on its platform “bots,” and for good reason. We always remember that we are speaking to a piece of software, and thus we can forgive it when it misinterprets our language or performs strangely. The bots can have strong personalities — the best ones do — but they don’t try and convince us that they are as human as we are. They know their functions and perform them well.

On the other hand, Facebook’s M attempts to combine human and machine intelligence into one personal assistant. When you interact with M, you’re never quite sure you’re speaking with a human or a bot. At this point, it’s also impossible to stop thinking about the distinction. M might solve the physical friction of actually performing your tasks, but this “Schrödinger’s Bot” approach introduces significant emotional and mental friction for the user. We’re naturally not used to the idea that something can be simultaneously human and inhuman. Covering up the mistakes of artificial intelligence with human intelligence only confuses the issue for the user.

Enhancement: Giving Our Humans Humanity

There is another way to strengthen the link between software and humanity. Instead of giving our software human-like traits, we can give the user superpowers that enhance their humanity, like increased empathy, perception, and creativity. Many messaging tools attempt this, but Snapchat more than any other social product gets this right.

By allowing our shared moments to disappear, Snapchat gives meaning and immediacy to our messages and forces the user to focus their attention on the present. User “stories” allow us to embody friends and strangers, taking on their viewpoints and forming empathic bonds. Live stories create shared experiences that are more than just the sum of our perspectives.

The landing screen for Snapchat is the camera. Anyone with the ability to press a single button can create content, and the product directly encourages this. The limited set of tools (pencils, lenses, text) on Snapchat allow us to be creative without overwhelming us. The strategy reminds me of one of my favorite Brian Eno quotes:

A studio is an absolute labyrinth of possibilities — this is why records take so long to make because there are millions of permutations of things you can do.
The most useful thing you can do is to get rid of some of those options before you start.

Snapchat allows us all to enter the studio, play with our identities and our appearances, and send them without fear of judgment to our network of close friends. This gives Snapchat a strong feeling of “realness” and authenticity. Which reminds me of another Brian Eno quote:

When people censor themselves they’re just as likely to get rid of the good bits as the bad bits.

Snapchat gets rid of unnecessary options and eliminates our need for censorship. As a result, it creates a deeply creative experience around making and sharing content. And most of the time, the content is human faces and bodies. Humans are naturally hard-wired to love looking at humans. Remember how Facebook started?

Speaking of Facebook, they provide a strong counterexample to Snapchat’s humanity. Algorithmically generated “Year in Review” stories approximate a human interaction — making a photo album for your friend — but very clearly fail. Instead of orienting the product around relationships and shared fleeting moments, the focus is on an algorithmic News Feed and profiles that never disappear. Instead of enhancing our empathy, Facebook creates echo chambers and amplifies our strongly held opinions. Facebook does not seem to understand the difference between a recommendation from a friend and an algorithmically-generated recommendation in our Feed.

This is not to say algorithmic approximation of human creativity is impossible or unwanted. Some software succeeds in delivering a positive user experience, like Spotify with Discover Weekly. And Snapchat is not the only company who succeeds: YouTube, Periscope and others have all found some success in bringing humanity to their product.

But if you ask me what the best human-oriented social software products in the marketplace are, my answer is unequivocally Snapchat and Slack. Both provide human experiences through endowment and enhancement. Both have friendly, grinning mascots. Both find ways to make collective intelligences and experiences that exceed the sum of the individuals, while still celebrating the uniqueness of our formed identities. And both have undergone tremendous recent growth.

And they’re not just riding a trend. They can potentially define it. When I ask my friends who the “human-first” tech company is, no one has an answer. This is a blank space in meaning that is waiting to be filled — and these are the two companies who can do it best.

Humanity can define your product, and it can define your culture. Humanity can be reflected in your leadership, in how employees are treated, in how you relate to your community, in how you talk about your product and act out its ideals. It can be the empathy you show to your users, the creativity you show in your design, the joy you take in creating.

Your humanity can and should be a competitive advantage. And if you choose to ignore the principles and examples above, soon there won’t be any humans using your product.

Don't Let Them "Gamify" Your Life

Many of the poorly designed games we play in life, such as work, need us to adapt them. Here’s how to think like a great game designer, not a lazy one.

Flickr -  Fabian Bromann

Is your life an RPG? Is it a first person shooter? Is it Solitaire, or more like Minesweeper?

It doesn’t really matter what you call the game of your life, or any game — just how well it’s been designed. Regardless of genre, all well-designed games have a central set of principles that work to create an engaging experience and help players achieve mastery. But those principles may not be what you think they are — in fact, you may be seeing everything backwards, like I was.

A couple of years ago, I got obsessed with the idea of “gamification.” As a lifelong gamer, I was intrigued by the idea that game principles could be applied to product design, business tactics, and even personal development. Who wouldn’t want to build a product that engaged and brought joy like their favorite games? Who wouldn’t want to achieve the high score in life?

Yet most of my attempts to introduce metrics and accountability — the “points, badges, and leaderboards” system of gamification — were complete disasters. I just ended up with broken products, broken promises, and some very negative scores. It turns out that although games are easy to pick up, many of them are just as easy to put down. People giving up on my products was one thing. But nothing crushed me like giving up on a game that I had designed for myself.

Go to the gym five days a week to earn the Super Strong badge! Nope, didn’t work. +1 to endurance for this long pointless meeting! Didn’t really motivate me. 100 points for taking out the garbage! Uh, it’s still sitting there.

Here’s the critical piece that turned it around: I had to stop thinking like a gamer, and start thinking like a game designer.

I read a paper recently that made this distinction very clear using the MDA model, explained below. I’ve adapted some of the images and points here, because I believe they tell the story well.

The MDA Model of Game Design

The game designer and the game player have a very clear relationship — one creates, and one consumes, like in the picture below.

Within the “blue box” of the game, there is a progression from the design of the game to the play of the game that creates an experience for the player. It can be broken down into MDA: Mechanics, Dynamics, and Aesthetics.

Mechanics are the actual actions taken in the game. For example, in chess, this is the movement of the pieces and the rules governing turn-taking and capturing. Bishops are constrained to diagonals, pawns move forward but capture diagonally, and so on. Mechanics can also set the win condition of the game, defining the capture of the opponent’s king as the end state.

Dynamics arise directly from the mechanics of the game as the game is played. In chess, the dynamics of aggressive gambits, defensive pawn structures, or time pressure (in blitz chess) are made possible by mechanics but are not directly codified in the rules. Over time, players learn to account for dynamics and use them to their own advantage. Experienced players are often very aware of complex game dynamics, while novices have just learned the mechanics and are simply struggling to follow the rules.

Aesthetics are the emotional experiences of the players that arise from the dynamics in play. Some common game emotions include surprise, fear, anticipation, jealousy, triumph, suspicion, and joy. Dynamics provide ample opportunity for players to experience positive and negative emotions in rapid succession. If players dwell in negativity too long without hope of triumph, they may give up on the game. But if they only experience success, they will grow frustrated with the lack of challenge.

The key here is that the gamer experiences aesthetics first and foremost when playing. The emotions are what resonate most strongly and what we carry with us after finishing a game, especially if we’re competitive (I am). If we fail, we rarely walk away cursing the rules — we curse our opponents.

But when a game designer fails, they don’t blame the players for not experiencing the right emotions. They blame themselves for not establishing the right rules. Without the right mechanics, complex dynamics don’t arise, and the intended aesthetic experiences are not felt.And usually game designers get very good feedback — if people stop playing or buying their game, they don’t make money. There’s no time to blame the player — only time to go back to the drawing board.

Thus designers and players sitting on the opposite sides of this model see the progression in reverse:

As a player of games, the mechanics are removed from me. I am taught them, but I do not control them. They are taken for granted as constraints on my environment. What I do experience directly are the aesthetics created by the dynamics, and these two things often take my focus.

But in the game of our lives, we are often the designer as well as the player. And if we get frustrated, many times it is the designer who is to blame, not the other players in our game.

A Poorly Designed Game: Work

Sometimes we are not the original designer of the games we play, and they may be very poorly designed. The game of work, especially for new players, might be the best known example: we are often given “badges” and “points” (titles and salaries) that are meant to demonstrate progress and provide a sense of mastery. We are placed in competition with our peers as though on a leaderboard. As we know, these mechanics rarely create dynamics that lead to positive aesthetic experiences for the “players” (real triumphs or surprising learnings). They’re just bad scorekeeping for a lazy game.

We have a few choices when the game of work fails to engage us: we can blame the other players (ugh, coworkers), we can blame the system (dammit, capitalism!), or we can stop playing (I quit my job and moved to Thailand!).

Or maybe we have another choice, one many of us make: we adapt the game. We introduce our own mechanics, constraints on our own behaviors that create new dynamics in the workplace that lead to the aesthetic experiences we crave. We set our own win conditions — “I will become a senior manager by age 30.” “I will lead this working group.” We devise our own rules and codes: “I will get all my work done one day early.” “I will not leave the company for four years.” “I will avoid Gloria.”

This is a healthy habit for creating a sense of autonomy and progression in the workplace, but sometimes we think like players, not designers. We generalize from positive aesthetic experiences without thinking about how they arose. “I just got a rush from thinking about my next promotion. Now it’s my new win condition.” “I just had a great meeting with Bill—now I’m going to try and join his department.” These can be dangerous rules setting us down paths that don’t actually lead to the emotions and achievements we want.

Instead, in adapting our game of work, we must think like game designers. What rules could reliably create the dynamics that lead to the emotions we want? One example:

“Okay, if I’m not allowed to backstab anyone at work or talk behind their back, then I eliminate the dynamics of gossip and politicization that create a toxic work environment that I don’t enjoy. I imagine others also do not prefer this environment. If I could somehow encourage everyone to play by this rule, we could eliminate these dynamics and establish a better aesthetic experience.”

In fact, this is exactly what good culture officers do at great companies — they design the game of work. They come up with the codes of conduct and manifestos that lead to the right dynamics that create the correct aesthetic experiences for workers. And if workers are unhappy, good culture officers don’t blame the workers — they blame themselves.

The   result   of a Google Image Search for “happy.”

The result of a Google Image Search for “happy.”

Generalizing Gameplay

Once I realized I was thinking too much like a gamer and not enough like a game designer, I couldn’t stop seeing the parallels. I saw how my family life, my health, my friendships were all affected by my failure of perspective. Thinking critically about mechanics, dynamics, and aesthetics — and how each arise — will help you improve the design of your life.

Mechanics: Which rules of my life are actually unbreakable rules? Which can be changed or adjusted? Which mechanics should I create because they give rise to the dynamics I enjoy, and should I ignore because they do not? Given this thinking, do I really need to maximize salary or achieve this title? (It could be yes.)

Dynamics: How do the dynamics I perceive in my life arise from the mechanics? How do the constraints I place on myself affect the way I interact with others? How do these dynamics create emotional experiences, such as stress or joy or fear?

Aesthetics: Which experiences do I truly enjoy and find valuable? What does achievement or mastery really feel like? Are others experiencing different emotions while seemingly playing the same games as me? How are their mechanics or dynamics different from mine?

We can all be better game players — learning mechanics and mastering dynamics — but we can also all be better game designers. In fact, the two skill sets are extremely related. Some of the best game players I know are the ones who have spent years creating their own games and generalizing their skills. And many of them also seem to be the happiest people I know.

So whenever you’re designing the game of your life or someone else’s — maybe as a parent, or as a boss — remember to think like the best game designers, not the laziest ones. Don’t just use points, badges and leaderboards to motivate. Remember mechanics, dynamics, and aesthetics — and remember how you want yourself and others to feel.

The best rules make the best games. Go out there and play.

The Power of Intention and the Future of Social

I realized something crucial recently - social media does nothing for me.

That doesn’t mean I don’t use it - if anything, I’m obsessed. I belong to countless groups on dozens of platforms, public and private. I’m plugged in 24-7-365 (and sometimes 366). But most forms of social media are designed for the short attention span, for quick inspiration, for unintentional use - that is, use without intent. When I say social media does nothing for me, I mean exactly that - it actively does not do anything for me.

Modern social media places a significant emphasis on discovery. The bored user, not knowing exactly what they want, is presented with a feed, or a timeline, or recommendations that guide their path through the product. And they’re always being prompted. Add your friends! Upload your pictures! Volunteer your data! Only then can the product give you what you really want to see - at least according to the product.

If the user only has a vague understanding of their goals, or only a few minutes to kill, this serves just fine. Sometimes a short video or status update is satisfactory. But most social networks simply serve our neomania - our obsession with the newest or latest information or media. There’s no great depth, only infinite breadth.

But what if I know exactly what I want? Exactly what I’m trying to accomplish? Here social products often fall short.

Social Networks are TV Channels, Not Tools

Let’s use myself as an example. I’m not running a client-based business. I’m not selling a product. I don’t want to primarily use social media as a marketing channel for myself or my interests. I don’t want to primarily use it as a soapbox or a platform. But most of the time, that’s the only utility I can see. And even if I wanted that, the products themselves don’t really tell me how to use them to achieve those goals. They function mainly as channels. Think of YouTube, Twitter, Facebook, Snapchat, or Instagram. The products themselves don’t really tell you how to build a following or how to create great content. And they don’t really care if you do - their primary purpose is to capture your attention for the longest possible time, and repackage that attention for advertisers.

So while it’s great at discovery, social media generally sucks at a couple things particularly - personal support and professional support. And as such, it doesn’t do anything for me - at least, it doesn’t fulfill my biggest needs.


What are my biggest needs really? Even if I were actively selling a product I wouldn’t call that desire to sell a “need" (though it could be a goal). Some of my core needs are personal, and some are professional, but they come down to the same underlying deeper needs - emotional support, a feeling of community, and providing for others.

If we want to establish and maintain communities based around providing and receiving emotional or professional support - communities that enable their members to grow and achieve - we see the power of intention is crucial.

You join a strong community based on intention. You communicate with other community members based on intention. You want to accomplish something, either alone or together. You want to develop an idea or change a mind, even your own. Intending to be someone’s “friend” on Facebook or “connection” on LinkedIn is not really enough - you just know you want to be associated with that person. But how many of your Facebook friends are true friends? How many of your followers would really follow you?

Community and communication are both derived from the same Latin origin - “communis”, or "common". Community is sustained and created through purposeful communication around common interests or goals. How much do you have in common with your high school friends that show up in your news feed? In contrast, how much do you have in common with your work friends or social clubs? Work and social life are communities. Facebook is a bunch of loosely-held acquaintances - on a good day.

So why is it that for so long workplace and social communication tools - intention-based social media - have lagged in performance and character? Why don’t we love those tools the way we seem to love discovery-based social media?

Well, the act of discovery itself drives a lot of our positive feelings towards social media. The idea that our next big idea, our newest friend, our favorite piece of content could be a single click or swipe away is extremely appealing. But really we (and psychologists) know that lasting meaning is developed over time - investing in a skill or community or friendship, reading a powerful book, taking a course, having a career.

Chat and Forums: Networks of Intent

The two most common forms of online interaction for intent-driven communities have been chat and forums. These tools have developed so little in the last twenty years in part because they are so successful. Chat is real-time and personal. Forums create a storehouse of community knowledge. Both can provide emotional support, build community, and allow users to help others.

If you want to really get something done, or really understand something, chances are you belong to a private chat group of experts (maybe on WhatsApp, or IRC if you’re a nerd like me) or an expert forum (perhaps reddit or Stack Exchange). You probably don’t sit on your news feed waiting for the right answers to roll around - and if you do, MY GOODNESS PLEASE STOP DOING THAT.

Those basic forum and chat tools haven’t really changed in decades, yet they retain huge communities and function fairly effectively. I’ve written about why before. But with the rise of new tools, could we finally be entering a golden age of social communication, collaboration and efficacy?

Take Slack’s insane growth rate over the past two years - the team building Slack as a professional productivity tool understands the importance of intention in social communication. The positive psychology of feeling useful to yourself and others comes from intentional, collaborative communication, and Slack’s integrations, automations, and design makes it powerfully productive. Slack took the extra step of adding personal touches like Slackbot for onboarding and GIPHY integration, mirroring the camaraderie and personality of friendships. As a result, Slack provides emotional support, builds community, and allows users to provide directly for others - and that’s the primary function of the product, not a side effect.

This productivity stands in stark contrast to the passive engagement of Facebook or Instagram. Overuse of social media - our leisure time - has been linked to depression, just like overwatching television. Meanwhile, people regularly report feeling most engaged and happy when fully active in the workplace. The real truth is that we need to work, and we need to feel good about our work, and we want to feel good while we work.

I’m done with the old social media. New social communication tools like Slack mix the positive features of social media (direct communication, beautiful design) with the most positive features of workplaces (productivity, integrations, and access to resources). As our social and working lives continue to interweave, there will be an increasing demand for social networks and tools that understand our need to be both personally and professionally fulfilled. That’s why non-professional Slack communities have begun to proliferate even though Slack is currently focused on enterprise.

We want tools that we control, not tools designed to control us. And we want - we need - to be as productive in our social communities as we are in our workplaces.


The future of communication is already here. Our new communities of intention are already being formed. Join us :) 


Exponential Growth Isn’t Cool. Combinatorial Explosion Is.

So much of the tech industry is obsessed with exponential growth. Anything linear is dying, or has been dead for years.

Moore’s Law had transistor density doubling every two years. Available bandwidth is increasing exponentially - perhaps 50% annually. Every digital product needs viral exponential growth - invite two friends who invite two friends and so on until you have seven billion active users. Everything is a hockey stick that you’re praying doesn’t turn into a parabola.

But looking for exponential growth just isn’t good enough anymore. It turns out that the things that grow exponentially can un-grow just as exponentially. Optimizing for exponential growth might mean sacrificing stickiness, user engagement, and satisfaction. You’re fighting churn and struggling to support your networks. And worst of all, exponential growth isn’t even the fastest growth there is.

Let’s take a quick look at our old friend linear growth. You know, y = mx. Slowly, over time, you have more of something than you did the day before. Your growth rate is a constant, never accelerating or decelerating. Comforting, familiar - but slow. Here it is:


But then you hear about exponential growth - y = m^x. Suddenly old dependable y = mx doesn’t seem so great in comparison. Exponential growth is seductive. It’s faster, bolder, and it looks way better on charts.

Just look at that green line. I don’t know what we ever saw in linear growth, to be honest.

But let us consider combinatorial explosion. This is a mathematical effect that occurs when combinations of variables or nodes within a system are each connected to another, or when branching factors are considered in search problems (like when analyzing potential chess positions x moves ahead of the current position). A common numerical example would be y = x! (the factorial of x).

This is not a very compelling or impressive effect for small numbers (or small time frames, if that’s our x-axis). Let’s compare exponential growth to combinatorial growth.

When x = 5:

  •      exponential growth: y = 2^x = 2^5 = 32
  •      combinatorial growth: y = x! = 1*2*3*4*5 = 120

However, for just slightly larger x, such as x = 10:

  •      exponential growth: y = 2^x = 2^10 = 1024
  •      combinatorial growth: y = x! = 1*2*3*4*5*6*7*8*9*10 = 3,628,800

And for x = 20:

  •      exponential growth: y = 2^x = 2^20 = 1,048,576
  •      combinatorial growth: y = x! = 1*2*3*4*5 = 2,432,902,008,176,640,000 (or 2 quintillion)

Wow, you may be thinking. A million of something isn’t cool. You know what’s cool? 2 quintillion. Forget exponential growth - combinatorial growth is sexy.

I have graphed the effect of combinatorial explosion slightly unscientifically below:

Um. Okay. Maybe it’s more dangerous than sexy. Like, too hot to handle.

Combinatorial explosion is not the kind of growth anything is equipped to handle. If this is searching branches in a chess position, or if it’s any number of issues in artificial intelligence, we’re looking at evaluation times that increase from seconds to millennia as complexity increases.

The reason for the huge disparity between growth rates is clear - instead of the base unit increasing exponentially, the number of possible paths or connections between base units is increasing exponentially. Big difference. That’s why we have developed methodologies like heuristic search to cut search times and better handle combinatorial explosion.

Our individual brains already work combinatorially. We have perhaps 100 billion neurons, and “each neuron may be connected to up to 10,000 other neurons, passing signals to each other via as many as 1,000 trillion synaptic connections, equivalent by some estimates to a computer with a 1 trillion bit per second processor.” Adding another 1 billion neurons wouldn’t be considered 1% growth - it’s a huge expansion in the number of possible connections and signals. 

Can Networks Explode?

Okay, so combinatorial growth is awesome, but overwhelming. So let’s say you’re focused on achieving normal hockey-stick, exponential adoption for your product. You definitely won’t see combinatorial explosion in your monthly active users or file uploads. Those kinds of metrics are going to stick to exponential growth - at best.

But what if instead of looking at the number of users, we looked at the efficacy of users on a social network? We can then see how the productive power of a social network can indeed increase combinatorially, and to our great benefit.

Before we do that, let’s look at Facebook, a network that has exhibited strong exponential growth even as it approaches the carrying capacity of the internet. That’s great, really. But if users only connect and converse with their existing real-life friends, no new connections or paths are formed. You may discover a friend of a friend on occasion, and you may add them as a friend but never directly converse. The productive power of the social network, or even your power within it, cannot be said to have increased since no new paths were added and no new communication was made. This is pretty realistic - Facebook as a product doesn’t encourage you to add strangers as friends or converse with them; it mostly funnels you into a news feed populated with lucrative advertising space.

LinkedIn does a little better - it encourages you to make connections with people you don’t know in real life or to refer your friends to new professional contacts. The focus on professional goals like networking and job-seeking is helpful here. Direct communication is limited though - either you use InMail (a slow, email-like messaging feature) or speak in LinkedIn groups (essentially old-school bulletin boards). In this case your productive power has increased somewhat, and the productive power of the overall network has increased as well with new paths established and additional communications sent. But the means of communication are so bad, your individual network’s power is still limited.



Now look at a chat-based productivity tool like Slack. Within a workplace environment, user growth and carrying capacity are determined by the number of employees in the workplace. We certainly can’t expect exponential growth in active users, and likely not in the total number of interactions. However, because network participants are organized and motivated by intent and the platform is built on real-time messaging, the number of possible connections and the frequency of connections can lead to a strongly productive network that continually generates new concepts, especially when properly managed and nurtured.

There’s a few additional advantages here. The creative power of the group is increased due to flexible group sizes - two users can add any other single user to a chat and have a materially different productive conversation. Also, the users themselves can seek out or break off conversations, thus helping to prune the explosion of new communications. Users determine whether ideas and strategies generated by these connections are likely to be useful to other users and can share as needed across the network.

Imagine if we combined the exponential growth of Slack users and groups with the combinatorial explosion of productive power enabled by individual Slack networks. Suddenly we start to see an entirely new paradigm for social communication that’s focused around productive generation of ideas, content, art and action across networks of engaged users. Different disciplines and expertise interbreed and birth revolutionary concepts. A beneficial combinatorial explosion of productivity and culture ensues! Rejoice, humanity!

Maybe that's overstated. But look, all Facebook’s exponential growth ever got me was a bunch of cats in my news feed. I never see those billions of users. As far as I know, their existence only helps Facebook optimize their advertising product. 

In my mind, it’s time to stop focusing on the hockey sticks, and keep our eyes on the puck. Social communication needs to move away from the echo chambers of Facebook and back into real-time, diverse networks enabled by modern, powerful, purposeful social tools.

So let’s start collaborating combinatorially. Let’s see what we build. Let's move fast and make things.

Playlists: How Listeners Take Ownership

Do you own Adele? Are you sure?

Music industry observers are holding up Adele's recent sales success as evidence that consumers still want to buy and own music in an increasingly on-demand digital age. For slightly more than the cost of one month on a premium streaming service - providing access to millions of tracks - consumers paid for access to just eleven songs. Unbelievable, right?

What does this say about the value of the long tail of music, or discovery mechanisms, or any individual artist? Unfortunately not much. For one thing, record sales are definitely not mutually exclusive with streaming service subscriptions. I am but one example of a consumer who maintains a premium music subscription and still purchases full albums. Secondly, Adele is an industry anomaly - you can't extrapolate from her success and draw any defensible conclusions about emerging artists, or even other megastars.

What it does say for certain is that listeners still enjoy feeling ownership of music. No, we didn't write Adele's songs, nor did we perform them (unless you're Adele reading this, in which case, thanks for your work!). But in addition to holding a recording, giving us control over how and when we play music, purchasing an album gives the consumer a feeling of access and partnership with the artist.

On-demand digital services, on the other hand, struggle to induce feelings of ownership. As long as you pay for your Netflix subscription, you have access to Netflix content, and that can feel like ownership (especially in hour eight of a binge-watch). But try viewing it offline, or after you cancel, or in non-covered countries, and suddenly any feeling of ownership of or access to House of Cards disappears. Ownership is an illusion that Netflix hardly tries to maintain.

So consumers clearly lack ownership in subscription models, but the reality is that we don't really even own downloaded digital purchases. Companies can take back our access to our "own files." Lose your hard drive, and you could lose your content. It is needlessly difficult to transfer some digital purchases between devices (especially if you used the iTunes store). Physical purchases are still okay as long as you physically hold them - but like any physical good, they are at risk of loss, theft, or destruction. You pay for the increased ownership, but maintain all the risk. You also need access to a physical player. My Macbook Pro is an incredibly powerful device, but it no longer has a CD drive.

So what really is "ownership" in music for the consumer? And how can subscription streaming companies give ownership to consumers?

The Playlist As User Generated Content

User generated content (UGC) is the key to the explosive success of many new-gen media platforms, including YouTube and Snapchat. Anyone can upload material to YouTube and immediately have it be searchable and potentially monetizable. Snapchat Live Stories are made entirely up of user-filmed videos and still photos, and are often much more engaging than professionally produced media.

User generated content provokes feelings of authenticity and builds a sense of community. For example, YouTube Gaming is entirely supported by tournament footage, Let's Play videos and game reviews uploaded by users. While this builds vast fortunes for a lucky few, for the most part players aren't uploading content for any explicit material gain. UGC is mainly created so that users can feel a sense of ownership and partnership with the community.

On platforms like Spotify, UGC manifests in the form of playlists. There are over 2 billion playlists on Spotify alone. While most listeners are not artists and do not have their own recorded music uploaded on Spotify, almost all of them use playlists to organize and curate their song collections - a necessity when you have access to millions of tracks. In this case, the UGC is being created for the direct benefit of the user who makes it. But it can also be used for the benefit of others.

Just like individual tracks, playlists are discoverable. Discovering a single track can be rewarding, especially for obsessives who can listen to one track for hours (Hello, that's me). But discovering a great playlist is even better. A well-maintained playlist changes and evolves over time, attracting great tunes to it like a magnet, paying musical dividends over weeks and months. 

Some playlist creators are major influencers with hundreds of thousands of followers, like Sean Parker. Some are brands or organizations, like the White House (above). But the playlists we value most are created by the radio stations we enjoy, or the artists we know and love, or our closest friends. These last playlists are the most authentic, made by our peers - but similar playlists are also made by the artists and curators we wish to feel closer to, building a sense of community around music. Importantly, playlists made by peers, artists, and Spotify themselves all look and feel identical in the product.

The vast success of Discover Weekly shows how an evolving, personalized playlist may be the best way of finding new music and inducing a feeling of ownership of music in users. In testing, personalizing the photo on Discover Weekly led to even more user uptake. That's Spotify's greatest trick: while users don't directly create the playlist, it still feels like theirs

Artists Embracing Playlisting

Of course the playlist is not a new concept. What is an album but an artist-created playlist of their own content? Isn't Discover Weekly like getting your favorite new 30-track album every Monday?

In fact, some artists are beginning to take the new possibilities of the playlist format even further than a static album. Matoma recently released a full-length LP on Spotify in the form of a playlist. The goal, he says, is to "update the playlist with a new single at least every three to four weeks for the next six months." And there's more possibilities too:

Say he wants to hold a special concert and within that set, he's going to launch his next single? Fans go nuts and want to hear the song again, and they can on their ride home when their excitement is at its peak, because it was just added to the playlist.

It's a fascinating move that feels more like the future of music than Adele's return to the past. While her success reminds us of the importance of quality musical content, Spotify's success reminds us of the importance of user-generated content in creating a sense of community, authenticity, and ownership around music.

Maybe instead of hiding playlisting functionality, other platforms can take a lesson. Let users take some ownership too.


My Discover Weekly: https://open.spotify.com/user/spotifydiscover/playlist/08j0RVCqhmp762ok6h2eNO

Ford's Finds (one of my favorite playlists): https://open.spotify.com/user/121815535/playlist/3CLr0P72igJNFJxnczJqQT

Augmented Reality is Already Here. It's Called Music.

Just when we were finally getting used to mobile, it feels like the next big thing in technology seeking to acquire and monopolize our attention is starting to arrive - virtual and augmented reality.

When we look at Facebook's Oculus VR, or Google's attempts with Glass and Magic Leap, or Microsoft's Minecraft Hololens demo, it's hard to ignore the future staring us in the face. Soon we won't be designing simple UX for flat, 5-inch displays, but the architecture of entire multiverses. We won't be checking in on Yelp or Foursquare - we'll be able to visit our favorite coffee shop without leaving the bedroom. Rich, immersive emotional experiences will be downloadable, transferrable, and pirateable. Physical space is being disrupted! Behold humanity's transcendence! 

Or maybe not. Maybe social norms are not ready to accept computers strapped to our faces. Maybe the technology will be prohibitively expensive, or perhaps it will remain accessible but low-quality. Maybe it will make some of us sick. And maybe, despite my personal optimism, reports claiming $150B in AR/VR market value by 2020 are a bit ambitious. 

But I am not here to praise AR/VR, nor am I here to bury mobile. No, I have come here on behalf of Music.

Despite the widely-reported death of the music industry, every time we dig it up we find it's still kicking. Adele sells 3 million records in less than a week. There are tens of millions of streaming audio subscribers worldwide, committed to paying every single month to access the world's music. It turns out killing an art that has existed for over 40,000 years - and perhaps far more - is difficult. How's that desktop computer industry doing?

It's natural for companies and consumers to get excited over emerging technologies like augmented reality. What if we had smart glasses constantly providing additional feedback and information on our surroundings? What if surprising creatures and characters popped into view behind everyday furniture, per Magic Leap's vision? How would this boost our productivity and our sense of wonder?

But this is what music already does. Augmenting our visual field is a new feat, but augmenting our sense of sound is one of the oldest human tricks. Listening to music helps surgeons perform better. Music surprises, inspires and thrills us through the release of powerful neurotransmitters. Music is what raises the hair on your neck in those horror movie scenes. Even the sound you can't hear can cause deep emotional reactions

Music is there when we seek to control our moods or immerse ourselves in our emotions. We inflict music upon ourselves, tying it permanently to places and people, providing sonic signposts in our memories of our lives. It's our breakup songs, our road trip soundtracks, our first dance. Sometimes music is unconscious; other times it's our entire consciousness.

One of the great challenges of AR and VR will be in providing an immersive social experience that mirrors that of music. When music plays in the background, it provides an emotional undertone for everyone in its presence. When Google Glass rests on your face, it is quite literally a barrier between you and the other person, introducing uncertainty and distance where music provides closeness. Are they filming me? What social metadata is layered over my facial features? It is difficult to imagine a first dance improved through the use of visual augmented reality. 

And yet it is commonplace - socially acceptable - to see humans wandering around in public, lost in thought, rocking out or self-medicating with music. If you ask Sony, the Walkman was one of the dominant cultural innovations in human history. If you ask Apple, it was the iPod. Either way, the emergence and acceptance of "personal stereos" was clearly a turning point for humanity's relationship with music.

Having music at our fingertips gives us something many of us lack but crave - control over our environment as well as control of our emotions. And thanks to mobile technology and music streaming platforms, we now have an entire universe of music available to us, instead of thirteen songs at a time.

Further augmenting its effect, the music we hear can now be responsive to our environments. Spotify allows users to discover popular tracks in their cities or even sync their music to their run. Technology also exists to sync music with gaming experiences or to soundtrack your surroundings.

Lean-back listeners don't have to know what they want to hear to enjoy themselves more than ever. Instead of a few radio stations available in each city, we have nearly unlimited personalized digital stations available to us at the touch of a button. Discovery has never been easier and reach never broader, giving hope to many thousands of aspiring musicians. For playlisters, soundtrackers, concert-goers and sharers, the availability and variety of music these days feels like painting with millions of colors instead of three stubby crayons.

Visual-based AR and VR are immersive and powerful, but they're fighting thousands of years of evolution to become our dominant emotional influencer. Music is already a deep, almost unconscious part of our lives. And how do you make the creation of powerful AR and VR experiences as democratic as the creation of music has become? When will infrastructure support the delivery of millions of high quality AR and VR experiences? 

It's not as though AR and VR could exist in the absence of music. Filmmakers are becoming the dominant creators of VR media, and they have some of the strongest appreciations for music. Remember that in film, music was originally considered more important than spoken dialogue. Judging from recent blockbuster releases, it certainly still is.

I believe in the industrial potential of AR and VR, and I buy into much of the hype, but ignoring or undervaluing music is a mistake. Music is what's worth investing in. You want to talk Net Present Value? The value of music spans millennia; media platform technologies cycle in decades. Scale? Music is absolutely universal, crossing every border and touching every culture; the Internet itself hasn't reached half the globe. Even Google may have realized that augmented reality solutions don't need a screen. They just need sound.

Why wait for the future? Playing music makes us happier and smarter. Hearing music makes us healthier.

Today, we already have music. And we always will.

YouTube Music: Google is Killing Itself

As a followup to my last post on Facebook’s push into music streaming, I thought I’d take a look at YouTube Music’s recent launch and Google’s push to make more sense of its disparate media-focused products as it tries to attract paying subscribers.

My takeaway: Google’s traditional "self-competition" strategy for launching and developing new products may doom its latest venture.

YouTube Music - Lifehacker.com

YouTube Music - Lifehacker.com


Google: Always Testing the Water

It is well known that Google has a love of testing and quantitative data in designing its products and services. (Who can forget the ever-popular “We Tested 41 Shades of Blue” story?) One of my favorite related theories is the “Google Makes Two Of Everything” theory. As this article notes, Google has a history of making simultaneous bets on competing products or technologies — Android Wear and Glass, Gmail and Inbox, Maps and Earth. Sometimes both products are a success (like Android and ChromeOS); sometimes the products are each failures (Google+ and Orkut…).

In the event where both products fail, it may be that Google’s lack of focus costs them substantially when they are up against an established, well-run competitor that is equally data focused. When it comes to social products, this was clearly the case — as Fortune explains, Facebook’s success essentially forced the dismantling of Google+. That same article points out some crucial lessons for entrepreneurs in the wake of the failure:

“Don’t sell people what you already have... Know who you are… Don’t compromise your core business.”

These are valuable lessons for startups and undercapitalized companies struggling to find product-market fit, but Google’s core business of search secures the company itself against any short-term failure. Each additional product launched through existing Google channels is an opportunity to gather data about their consumers — meaning they can take a longer view on product design and market success.

But this is not always a winning strategy. If the products being designed are targeted at consumers whose needs have long been underserved, Google can afford to take its time in developing strong products and business models. When it comes to social, however, Facebook simply launches better products and solutions faster for users who want to share, follow, and connect. They’ve learned from Google’s example on data-driven methods — and from Google’s ambition. Now Google’s failure on social products has led to Facebook finding a foothold, and their success may cost Google its core businesses of advertising and search.


Google’s Messy Media Products

Despite Lifehacker’s valiant effort, splitting Google’s media product offerings into just two competing services (Music+Red vs. Google Play) isn’t even realistic. Beyond that, it’s not clear from Google’s marketing which services you get from paying for which product. As of now, downloading YouTube Music gets you a free two-week trial to YouTube Red, which gets you ad-free playback and offline listening, but only for videos, and Red comes bundled with Google Play, but Play is also bundled with YouTube Red. This all replaces YouTube Music Key, but Red isn’t its own standalone app, while YouTube Music is, and then there’s YouTube Gaming…

(I had severe difficulty writing that paragraph, so it’s very understandable if you had trouble reading it and understanding it.)

The point is, Google is asking consumers to give up other streaming music subscriptions to come try out an ostensibly new service, for only a two-week trial (versus Apple Music’s 3-month offer), on a platform that really only has utility if you’re paying for the underlying YouTube Red service. Using YouTube Music without Red means no audio-only mode, no offline mixtapes, no features that actually make it feel like a music product — or at least not one positively differentiated from Google’s existing offerings. YouTube Red itself is kind of a mess, due to the lack of a standalone app and an inability to host some important content on its no-ad subscription-based tier.

With music, Google has stuck to its existing strategy of launching competing solutions and learning from the resulting feedback and data. As TechCrunch noted in its initial YouTube Red launch story:

The service makes a lot more sense than Google having a slew of limited subscription services with restrictions as to what you could access. The company says that watching the YouTube Music Key beta, it learned that users didn’t want to be told what was considered “music” and would be ad-free, and what wasn’t.

Wonderful. But should it really have taken Google one failed product and a long-term beta to reach this insight? And why didn’t they consider the cost of confusing their users in the meantime?

Made in Google’s Image — and Now Beating Them

While Google set the standard for the 21st century, they seem to have underestimated the competition. The landscape surrounding Google is not what it was for the first decade or so of this millennium. Companies like Spotify and Netflix (in addition to Facebook) have heavily data-informed processes and strong independent brands in the media world. Consider Spotify’s investment in analytics powerhouses and Netflix’s commitment to recommendation. What was once Google’s advantage is now the industry standard, and its weakness — coherence in product and marketing — is being exploited.

Google moves methodically, as befitting a 60,000 person company, but media is evolving faster than ever before. Mobile is already dominant, AR and VR are pushing into the consumer market, old players are dying and mobile-only ad-supported media platforms are finding huge valuations (see: Snapchat).

Spotify and Netflix are demonstrably nimbler, and they’ve already solved one of media’s biggest challenges: extracting value from the consumer. And with their strong product features and excellent discovery and recommendation engines, they are providing more value back to the consumer for their time spent on the service. While YouTube still wins on community, Facebook is challenging Google directly on social and now video as well.

The competition is mounting for Google, and its old methods of self-competition and disruption won’t work in this new, more intelligent tech media landscape. The time would be now for a coherent, standalone streaming music offering — but YouTube Music isn’t it.

Forget YouTube. Facebook May Win Music Streaming by Leveraging Social

In the past few years, music streaming has suddenly become one of the most crowded spaces in tech. Established players such as Spotify and Pandora are fighting off challenges from some of the biggest tech players in the world. Apple Music launched with 11M users in its first month, Amazon Prime Music - visually a Spotify clone - comes at a great price point for Prime subscribers (free), and now YouTube Music is launching as a standalone app. Even this ignores other strong players in the space with large non-US user bases, such as LINE and Deezer.

Arguments have been made for each of these platforms "winning streaming," mostly focusing on things like Apple's dominance of mobile hardware, YouTube's ubiquity and free features, Spotify's analytical intelligence and strong product design. But what does "winning" entail? Is it being the largest platform, having the strongest product, having the most momentum?

For the moment, this space is significantly fragmented and consumers often use multiple music solutions, showing loyalty to none. Each company uses different channels to promote their product (Spotify and telco deals, Apple and iOS devices) and differ in strength on mobile. This creates an opportunity for any of them - and perhaps the largest opportunity for one of the only major tech players without a standalone music service. Facebook.

Facebook also had a major product announcement recently - "Music Stories", a feature launching initially to Facebook's iOS app users. The feature allows users to play songs shared directly from Spotify and Apple Music on their news feeds and instantly add those songs to their music libraries on those services. For the moment, users can't stream more than short 30 second samples on Facebook, and those are the only two services supported at this time. That's a modest start for Facebook, which has previously supported Spotify integration.

Why is this Facebook's play? For instance, why haven't they yet launched a standalone music service? Signs already point to Facebook making an even larger push into music, and they have already invested heavily in the success of their video product. I'll examine what Facebook might be doing, and who they might have learned from.

How To Fail At Music Streaming

We get a better idea of Facebook's motives when we look at one competitor in particular: Apple Music. Launching in July, Apple Music boasted an impressive 11 million listeners in its first month of availability, but by October had only added 4 million more. 6.5 million of these listeners were paying for the product, leaving the remainder (8.5 million) as free-trial users. While this may initially seem like a decent retention rate, user growth was clearly awful in the intervening months, especially considering the massive amount of available iOS devices that could simply sign up for a FREE TRIAL.

So what happened? Apple relied on a massive initial marketing push, which briefly boosted awareness of the product, but neglected to add any strong social features at launch. Without the virality enabled by direct sharing of songs, Apple Music relied on word of mouth after launch to attract users - and often that word of mouth was not good.

What happens when products neglect virality as a growth mechanism? They don't attract new users, and then they die. In fact, Apple Music did their competitors a favor by educating users about the ease of streaming music and failing to capture any of them. The biggest reason: there were no significant network effects associated with use of the Apple Music product. With no free tier on Apple Music to retain users who were on the fence, Spotify and YouTube were the beneficiaries of new music streamers who were unwilling to subscribe to Apple Music after their initial trial period subsided.

I used Apple's own published numbers on conversion rate and user growth to project their growth based on a virality model (see Andrew Chen's post). If we attribute every single new user to viral growth and reverse engineer the viral model from their numbers, each Apple Music user was responsible for attracting 0.3 additional users to the platform (a high number, considering many new users simply discovered the app through updating their OS). If we keep the retention rate constant (and the overwhelming evidence suggests it may decrease), Apple Music's organic viral growth should flatline around month 5. Which is to say - now. And that's just for free trial uptake.

What may boost their success above this organic line might be additional marketing efforts, new iOS users, expanding to Android platforms, or actually building strong user-to-user social sharing features, but the damage is already done. 11 million users in the first month were dumped into a terrible funnel, and most have already leaked out to competitors. Apple blew it.

Apple Music's hockey stick seems to have fallen down.

Apple Music's hockey stick seems to have fallen down.

These abysmal growth numbers are corroborated by Google Search trends, which normally does a pretty good job of mirroring things like app adoption.

Yes, Apple Music is the red line.

Yes, Apple Music is the red line.

The Opposite of Apple - Embracing Social

So if you're a consumer who's finally willing to pay for streaming, who do you use? It may be the cheapest platform, or the one you are most familiar with, or perhaps the one that already integrates with services you use - this is YouTube's hope with including their premium music product bundled with YouTube Red. The one place you clearly won't go? Facebook, because they don't have a complete music streaming solution. No full songs, no expansive library, no discovery tools. So how does Facebook get started with launching its own music streaming solution? By attacking other services where they are weakest, and where they are strongest - social.

Facebook is unlikely to want to compete on price - undercutting the market and providing extensive free trials doesn't work as a standalone strategy, as Apple proved. It could compete on product, but a superior product does not necessarily guarantee consumer adoption.

Consider the following model, which I have adapted from SInan Aral's lecture notes. (Sinan is an MIT Sloan professor and researcher focusing on product virality.)

Product A launches at time 0 with an inherent value of a. Product B will launch at time t in the future with an inherent value of b, where b is greater than a. Both products add users at a rate of one user per time period, meaning product A will have t users at time t, while product B will have zero. Each product exhibits network effects, and the two products are substitutes, not complements. This network effect, c, is multiplied by the number of users of the product and added to the product's inherent value. The value of each product over time is shown below.

Image adapted from slides by Sinan Aral, MIT Sloan professor

Image adapted from slides by Sinan Aral, MIT Sloan professor

A new user entering at time t will see the marketplace as described above: product A's value is superior to that of product B, even if product B is inherently more valuable or complete. This new entrant will thus choose product A, especially if influenced by his or her close friends (who may already be users of A).

Imagine product A here is Spotify, which gains in utility as more users join the service. It has a substantial amount of data at its disposal and a large amount of fluency when applying the data towards improving their product (thanks in part to its acquisition of the Echo Nest). It also has a nominal amount of social features, though they are poorly designed and difficult to find. Even if a superior product launches at a later date, new entrants to the marketplace are likely to choose Spotify due to network effects. In addition, existing Spotify users are likely to continue using the product (especially with new personalized features such as Discover Weekly).

However, imagine a different product B - one that builds upon the existing user base of product A and functions as a complement, not a substitute. Now at time t, we see a different marketplace:

Image adapted from slides by Sinan Aral, MIT Sloan professor

Image adapted from slides by Sinan Aral, MIT Sloan professor

Clearly the math has changed for the new entrant. B now looks more attractive due to its higher inherent value as a product and can capture new users. My intention here is obvious - Facebook is attempting to be product B and avoid the fate of product A. And of course, Facebook has its own social graph to leverage in addition to that of the products they're piggybacking upon.

Facebook must see an opportunity to be the missing social component for all music streaming products and thus leverage the total cross-platform streaming user base instead of first attempting to dump a complete solution on its captive user base. Apple attempted to use its captive hardware users to expand its own music platform - it's not a winning strategy.

Why else might Facebook do this? By encouraging users to share their musical tastes on Facebook, they capture data on the musical tastes of existing streamers BEFORE launching a standalone product. This allows for greater personalization and overall user intelligence.

Facebook is attempting to construct a habit loop for users, and the video at the top of the article demonstrates this clearly. Use music streaming app. Copy song link. Go to Facebook. Share. What is the key to this loop? Leave your app, and come to ours, because we have the best sharing functionality.

This works because music is inherently a social experience, one shared between listeners in addition to between artist and fan. Apple overinvested in the latter relationship; Spotify initially leveraged the former but has lost its social strength as the product grew in complexity. Now Facebook is building the streamlined social music sharing feature Spotify should have had - and as a result, they are building a core user base conditioned to share and discover music outside of their current music streaming app of choice. Then, when Facebook finally launches its own standalone music streaming solution, it has already introduced network effects and a habit loop for users.

And crucially, this sharing functionality is free. This means they attract the widest possible user base to begin with, not just a subset of already engaged premium music streamers, who may be the hardest to initially pry away from existing services.

As we've seen, Facebook is learning from its past relationship with Spotify and from the failures of its competitors. But any number of things could affect the future growth of their music products, from the quality of the eventual product to market positioning and price point. How long it takes Facebook to launch their own music streaming product will also be crucial - wait too long, and they risk losing users to the competitor services it is promoting in the News Feed. Move too quickly, and they won't reap the benefits of their current partnership strategy.

And of course, competitors aren't standing still. YouTube Music is just getting started. Apple has billions in the bank. Spotify is still growing like crazy and adding video and podcast support. The list goes on.

So grab your popcorn and enjoy the plethora of amazing musical solutions we currently have as consumers. The battle is just beginning.

Closed Consumer Systems, or Why Apple Still Makes Maps

This post is about the value of consumer data to corporations, as seen partially through the lens of systems theory. I try to explore exactly how companies draw you into their product web - and why they really can't afford for you to leave. I’ll also try to show why these same principles explain why most recommendation systems will never truly enable discovery on the part of the user.

First, some background on the value of data and data analytics. According to IDC, the worldwide big data analytics market will reach $125 billion in 2015. Companies more than ever see analytics as a critical priority for their industries. Analytics is now a must-have across the board - a baseline, not a competitive advantage.

For those newer to analytics, a few definitions are helpful. There is descriptive or exploratory analytics, where underlying data is visualized or filtered in such a way as to highlight areas for attention and make insights more easily discoverable. There is predictive analytics, where data is used to generate models that can project with some confidence key dependent variables, such as future performance, demand, cash flow, and so on. Then there is prescriptive analytics, where models and data are used to generate a recommended business action that will optimize some key metric for the company. See more here.

When discussed inside corporations or between shareholders, the value of data is often stated in terms of the data itself. How comprehensive is the data? How many users or days do we have data for? How reliable is the data? How clean is it? How relational is it? How easy is it for us to work with? Who else might value it?

However, the real value of internal data to companies is really only in how predictive and prescriptive the data can be, especially for B2C companies. Is the data complete enough to form a reliable model of consumer behavior? Is it relational enough to identify key drivers of important user performance metrics?

Facebook, Google, and other large tech companies with a variety of product offerings already have enough user data to perform a variety of seemingly impossible tasks, such as tracking flu outbreaks or influencing our mood. Thus the key for them in leveraging our data (yes, OUR data) is not collecting more of it - it’s creating environments in which the data they have can serve as a complete model for our behavior. And that’s where it gets scary.


The Race to Close The System

I’ll digress for a moment to discuss some definitions in systems theory. Bertalanffy describes an open system as one where there is "an exchange of matter with [the] environment, presenting import and export, building-up and breaking-down of its material components.” In other words, there are external forces acting upon the system and driving behavior of the system’s participants. In contrast, a closed system is one characterized by isolation of its elements from external forces.

When you attempt to model the behavior of an actor in a system, you rely upon two things chiefly: data and relationships. Data can describe the amount of something, such as the number of users, or the rate of something, such as the churn rate or subscription rate of users. Relationships can show how these amounts or rates change over time based on their relationships to other amounts or rates. For example, churn rate may be linked to the availability of competing products, or subscription rate may be linked to marketing spending in certain channels.

In open systems, much of the data or relational knowledge required to construct an effective model is unknown or unreliable. For example, it is difficult to create a model that takes as an input a competitor’s marketing spend if the competitor closely guards that information. Thus consumers can appear to behaving randomly even when there is a clear but unmeasurable external stimulus. Tracking down these exterior forces is costly, often prohibitively so. Generally they are only uncovered through extensive A/B tests, which can still be inconclusive, or focus group testing, which is inefficient. And even these insights would still need to be translated into business action (prescriptive analytics).

Additionally, when a customer leaves a particular service, their behavior is no longer measured by that service. Therefore there is no way to close the feedback loop with a clear understanding of a consumers next behavior. All you know is that they abandoned the service - not why, or for whom. That means forming a prescriptive model is even more challenging, because it cannot always learn from the results of actions that are taken.

This is why large tech companies with a trove of data will often go out of their way to create closed systems for their users. The more data that is held internally and the more relational the data, the better the predictive models that are generated. But additionally, in closed systems, tech companies can artificially restrict stimuli so as to remove variability in user behavior. This makes it easier to predict purchasing activity or project the lifetime value of a customer - key business metrics.


The Only Winning Recommendation Is Not To Play

See, Amazon doesn’t care about recommending you the product you’re going to enjoy. It cares about showing you the product that you’re the most likely to buy, because that’s what it can predict. And then Amazon will spend all its time convincing you that the product you’re the most likely to buy is the one you really wanted all along.

Netflix doesn’t really care about allowing users to truly “discover” content. It will recommend you the movie you are the most likely to watch next, because that is what it can predict. And it will try to convince you that your taste ought to be more narrow than it is (“Teen Paranormal Dramas”, anyone?) because that will make its recommendation system look more powerful.

And most music services don’t really care about providing a good “discovery” mechanism for new music, because that’s not what they want either. They want to deliver a certain number of listens to the labels with which they have relationships, so they can continue those relationships. They want to program content such that the artists that are supposed to get listened to actually do. How can music platforms negotiate the value of content with content providers if they don’t understand the behavior of their own audience and their relationship with that content? Then that question quickly becomes: how can they understand and predict their own audience unless they control their users' behavior?

All of these companies create habit loops for consumers, and they’ve gotten really goddamn good at it. They can identify and test our triggers because their databases are full of correlations, if not causations. They do this not to keep us satisfied - they do it to keep us predictable. They do it to keep us monetizable. And there is very little business incentive for most companies to structure products or ecosystems any other way.

So the next time you use Apple Maps - and it still sucks - just remember why it exists. Remember that companies don’t want you to discover anything new, to find surprising content, to explore. Those are human needs - not a corporate need. So please, fight the habit loops and the seamless systems, because you know they won’t do it for you.

Whose Story Is It, Anyway?

I’ve posted elsewhere before about Snapchat's Our Story feature and how it enables complex storytelling and creativity. By allowing users to tag their Story submissions, multiple perspectives can be synthesized into unique coverage of events and locations. It’s powerful - but it’s flawed.

Of course, anyone who follows an Our Story only sees a curated selection of snaps. If you film yourself at a particular event and tag your Story submission, you’re most likely not going to make it into the public feed.

If you want to be part of the story, you’ll need to meet at least one of the criteria below (note that I’ve developed this list empirically by watching lots of Our Stories):

  • Be young and very attractive.
  • Be located behind or near a celebrity / sports figure.
  • Be extremely energetic while filming the climax of an event.

Unsurprising, but it’s clear that this is not an exhaustive list of perspectives. The fact is that Our Story is really just one person’s story: whoever is curating the feed for the sake of the event narrative.

So who is that person?

According to the job description posted by Snapchat for a Content Analyst, it’s someone who is:

  • A passionate storyteller
  • Familiar with music artists, trends, internet memes, and popular slang
  • An advocate for free expression who isn’t easily offended
  • Has the ability to remain objective and neutral while evaluating potentially controversial content

The third bullet intrigues me most. The way I understand things, it’s only important that the content analyst is an advocate for free expression in so much as it enables them to do their job effectively. When creating Our Stories, they aren’t expected to be impartial or omniscient. They just need to be tolerant.

There’s an artistic analogy here that helps me visualize what’s really happening. The Content Analyst’s work is much like that of a painter who uses a palette comprised of crowdsourced faces, emotions, and movements to depict a particular microcosm of culture.

And we, the users, must remember that we do not hold the brush or even see the canvas.We are simply the colors.

To continue the metaphor: it makes sense that the colors that get chosen are the most vibrant, or the most beautiful, or perhaps above all the ones that fit well next to each other.

As they are currently produced, Our Stories do not have room for dissension, or conversation and response, or debate. They’re not Hardball; They’re the Truman Show with a single producer and millions of actors, all desperate for screen time. They are just one person’s story told by many faces.

Our Story can be so much more.

Our Story is clever, and it’s monetizable. Smiling young people and unfiltered video build trust in the viewer. Peeking in on a Story is a very different experience than watching a promotional short or traditional news media coverage. It feels honest, even when it’s clearly marked as sponsorship.

Our Story has one strong limitation - real narratives do not have neat beginnings, middles, and ends. They are continuous and they are evolving and they do not lend themselves to summarization. The centralized curation approach to Our Stories ignores this for the sake of creating something more interpretable and palatable - for obvious marketability reasons.

But I see a power in this medium beyond what can generate revenue for the platform. It is the power to give a real face and voice to the public - and not just what a Content Analyst chooses to show. It is the power to create something very honest. Snapchat’s platform already allows us to create content that is visually striking, native to mobile, and easy to share.

For those people who are not beautiful vibrant colors, most do not have the means or desire to maintain a YouTube channel or Twitter presence. But some of them have waited their whole lives to tell their stories, and they know that no amount of Content Analysts are going to accurately portray their version of reality. They will, as always, have to create that story for themselves.

So what versions of Our Story might actually be our story?

I don’t think anyone has a complete solution. Snapchat has shown a great willingness to test and iterate, so I’d like to see them use this portion of their product to try other approaches to collaborative content. Some of my brainstorms:

  • Crowd-curated: top submissions are chosen by popular vote or chosen by users that are elected by popular vote. (This might be most useful for campuses, workplaces, and urban activists.) These submissions / users reset on a regular basis.
  • Unfiltered: all tagged submissions are entered into a continuous feed automatically that maxes out at a given length. Submissions reaching a certain number of total views or downvotes (left-swipes?) are removed form the feed.
  • Responsive: different users are shown different versions of Our Story dependent on location, previous interactions, friends submitting similarly tagged content, etc. (This possibility fascinates me completely and, for all I know, has already been implemented to some extent.)
  • Independent channel: a single entity conducts individual interviews to crowdsource perspectives (like many media channels such as Verge are doing on Snapchat).

To again state the obvious, Snapchat is young - it was founded seven years after Facebook and five years after Twitter, and it’s just starting to find a fit in a changing media and communication landscape. This isn’t supposed to be solved yet.

Then again, solving Our Story is crucial. Snapchat’s power has always had roots in its feeling of authenticity. Our friends don’t touch themselves up or self-edit and they feel free to express themselves fully knowing their emotions and perspectives don’t stay on their permanent record.

With the right approach, Our Story can build on this authenticity instead of undermining it. But only if it’s truly Ours - not Theirs.

Three Ways to Escape the Echo Chamber - And Why We Must

Social media is, for many, an escape. To glance down at our phones is to enter a safe world composed of our friends and family sharing our preferred content. This allows us to maintain the strength of our relationships even from a distance.

These strong ties also fashion a tough, protective cocoon out of our newsfeeds and notifications. Recommendation algorithms feed us the most relevant content to our interests, or the content deemed most likely to engage us. These algorithms are developed based on our observed actions and histories - not on our thirst for new experience and strange knowledge.

This is an inevitable result of the way these social systems are designed. On Facebook, the quickest path to content is the easiest, and the most likely to delight us: through our immediate friends. But the easy path also lulls us into a passive relationship with social media - we see the same websites shared over and over or small variations on a headline theme. With research showing that people choose to associate very strongly with those of similar political persuasion, it’s likely we only hear or see those views we agree with. Thus the echo chamber is created.

It’s important to note that we do this by choice - but the choice is passive. Because we long ago picked our preferred friends, we long ago committed to the content prioritized on our news feed. Because we long ago performed certain actions online, certain advertisers are shown to us that reflect our entrenched interests and perspectives.

But in my view, a truly healthy relationship with social media means actively using it to find content that can expand us, not simply sustain us.

I have found a few reliable ways to escape the echo chamber - to have a sort of social media “out of body experience.” I use these tools daily to step into the minds and perspectives of those unlike me. There’s two reasons I absolutely must do this:

  • I want to make things better. I believe that to change the world, we must first seek to understand it and love it. To hear others’ thoughts in their own words and to visualize their perspectives humanizes them and expands my knowledge of them and their worlds. Social media can be used to build understanding and empathy - and without them, nothing can change for the better.
  • I want to stay happy. Reading things that could upset me or that I could disagree with is sometimes unpleasant, but always necessary to sustain long-term happiness on social media. To maintain a healthy (read: not depressing) relationship with social media, it has been shown that you need to remain active in using it. That means engaging directly with other users by commenting or creating your own content to share on social media, rather than simply using it for news aggregation. To me, it also means being active in finding the content I consume on a daily basis and stepping outside the “feed.”

Here are some simple methods I use every day to escape the echo chamber. I hope they work as well for you.

1) Curate lists of alternative opinions.

I have often used the lists feature on Twitter liberally to organize the voices my life - friends, coworkers, influencers, trusted news sources, comedians. But one list I make sure to check daily is my “Disagree” list. It is entirely composed of people I generally do not see eye-to-eye with: mostly certain columnists, politicians, and businesspeople.

I’ll take some time to go through and see what content they’ve created to share, what issues they are engaging with, how people are responding to their message, and so on. Reading through gets me out of my comfort zone and helps stimulate my creative mind. It also helps me understand the language these influencers use and the effect it has on those who follow them. Being active in looking through this list is important - even when the content is unpleasant, I’m able to channel negative feelings into strong, productive energy afterward.

2) Visit pages for things that I don’t use or follow.

On Facebook in particular I like to visit brand pages for products I don’t currently or would never use. I ask myself: How is the company trying to reach its customer base, and who do they believe their customers are? How is it different from the way brands normally try to engage me, or people like me?

I’ll also visit fan pages for artists or shows that I don’t have interest in. Then I ask: how do the fans interact with their interests? How is it different from the way I choose to interact with my interests? How do the artists or producers respond? Understanding other peoples’ communities helps shed light on the way I form and build friendships and how I participate in my interests.

3) Use the “Discover” feature as the main feature.

I believe the Discover feature is not meant to be a novelty on social media - it’s supposed to be the primary feature. On Twitter, I frequently go through the top trending hashtags and choose one that seems irrelevant to my interests. I read manually through recent tweets and figure out for myself what’s getting talked about and why.

Meanwhile, on Snapchat, I use Discover to briefly scroll through content from brands I don’t particularly care about. I follow promoted Stories from events that I wouldn’t particularly want to attend. I watch the faces of the people involved and I recognize the joy and excitement that I feel toward my own strong interests and close friends.

I strongly recommend doing these three things every day, or as regularly as you use social media. If you use these methods, I promise you will end up with a stronger and more active relationship with social media. You will also be happier and more understanding of others - and maybe you will discover something crucial that you will carry the rest of your life.

Remember: Social media is a very powerful tool when we wield it, but it is just as powerful when it is wielding us.

You are not your feed. Go explore.

The Uncanny Valley of Social Media


How does reading that make you feel? Good? Bad? Pleased? Uneasy? Physically ill? Some young users see the word “Facebook” and feel revulsion. Their trust as users has eroded.

I'll take us through Facebook’s past to the evolving fields of animation, robotics, and virtual reality as I attempt to address a common question: why does using Facebook feel so unnatural?

But I’ll start by answering this question: did it always?

Famously created in a Harvard dorm room, Facebook has grown to be the dominant medium for online social interaction among young American adults. At its inception, Facebook focused on serving the isolated Harvard community. If you joined the site and added a friend, chances were strong you already knew the other person by name or face and that you had multiple mutual “strong-tie” connections.

A dramatic rendition of a Harvard dorm room.

A dramatic rendition of a Harvard dorm room.

As Facebook slowly expanded, it allowed other college students in the Boston area to participate in its network. Then Ivy League and Stanford students were invited. From these beginnings Facebook grew to the behemoth it is today, now having over 1 billion monthly active users and a strengthening mobile presence.

Demand for Facebook’s growth also meant demand for Facebook to change. In one of the original versions of the site, users simply maintained a personal page that friends could discover and interact with. As features were added, users at first delighted in their new capabilities. Suddenly you could post on walls, invite friends to events, and create groups. Now we can browse company pages and news feeds, play games hosted on the Facebook platform, and yes, still “poke.”

Somewhere along the line, though, we began to feel betrayed.

Multiple placed ads interfere with our engagement or disquiet us with their specificity. Fake profiles and spammers demand our attention. And recently, users were outraged by Facebook’s willingness – perhaps eagerness – to manipulate their emotions by altering the content of their news feeds.

We used to love the freedom Facebook afforded us in the online social experience. We used the service gladly, and so did all our friends. We had nothing but good will. Facebook simply wanted to approximate and augment our real-life social interactions so that we could take our existing relationships into the evolving online world.

So why did it all go wrong?

The answer is simple – it had to.

What do I mean? To answer this, let’s turn away from social media and toward the fields of robotics and animation.

As our technological capabilities have increased over time, our ability to approximate the human form using computers has grown dramatically. Robots have evolved from simplistic humanoids to near-lifelike representations of actual humans. Animations, once grounded in the two-dimensional plane of cartoons, have grown so complex and commonplace as to appear in every summer blockbuster. Visual and digital effects artists earn – or should earn – top billing alongside the stars.

So if technology has come so far, why are we still so repulsed by videos like this?

In 1970, the Japanese roboticist Masahiro Mori coined the phrase “uncanny valley” to explain our complex empathic response to human-like robots. His hypothesis was that a human observer’s emotional response to a robot was closely related to the robot’s appearance: as it grew more human-like, observers became more positive and empathic towards it.

However, Mori also explained the existence of a critical point beyond which observers were actually strongly repelled by the appearance of the robot. As robots come to resemble a human almost perfectly, a feeling of not-quite-rightness overcomes the observer and causes a powerful negative reaction. The feeling is so strong that any productive interaction between the human and robot is eliminated.

This is the uncanny valley – the chasm separating useful tools of human representation and actual humans. And in this valley, we feel lost.

The Uncanny Valley - credit in image

The Uncanny Valley - credit in image

Let’s climb out of this valley and return to Facebook.

Recall how the service has evolved. At first your Facebook profile was an imperfect approximation of your social identity. A static photograph of your face and a few lines of text summarizing your beliefs served to encapsulate your existence. No one could confuse your Facebook profile with your actual social persona – the complexity required to represent it did not exist.

As features were added over time, our profiles became part of our identities. We posted photo albums, added “life events,” checked in at landmarks, and shared deeply personal status updates. Instead of simply mirroring our rich social lives on Facebook, we lived fully online. Our online interactions were no longer merely reflections of our real-world friendships. We used them to create our perceptions of ourselves and others. Our digital selves became extensions of our physical selves.

So as our personas and our technologies have evolved – as Facebook’s services evolve to allow for more complex online interactions – we have gained the ability to nearly approximate our real-world relationships on the Internet, such that we can almost seamlessly flow between our online and physical worlds.

But it’s not quite perfect. And there is the problem.

My perception is that we are in the uncanny valley of social interaction online. We’re close to something good – close enough to know that we are close. But also close enough to know that something is terribly wrong.

There is still something about the way we conduct ourselves online – whether our platform is Facebook, Snapchat, Twitter, forums, comment sections, and so on – that is perverting our social interactions and our perceptions of each other and ourselves. These services allow us to be human-like, sharing our ideas and faces and actions, but we are still forced to be distinctly non-human in a way that repulses us. All this is happening while we struggle to learn how to absorb our online personas into our identities.

Let’s be clear. The problem isn’t the users. It’s not the fault of a particular generation or group of narcissistsThe problem is the systems themselves – not just in the way they’ve been commercialized, but also in the way they’ve attempted to accurately replicate, augment, and replace our social interactions while integrating new aspects of our digital personalities.

As investors demand growth and users demand new services, our social platforms try to do too much. In pursuing the essence of our social interactions, the platforms are coming closer, but they’re still not quite right.

And in the meantime, we won’t feel quite right either.