Microsoft and Apple Double Down

Posted on

It has been years since Microsoft upstaged an Apple Keynote (such things usually run in the opposite direction), but that is exactly what happened yesterday with the former’s $26.2 billion purchase of LinkedIn overshadowing Apple’s impressive yet iterative announcements at their annual Worldwide Developer’s Conference (WWDC). And yet, despite the contrast in expectations (zero versus high) and reactions (“What?!?” versus “Makes Sense”) there is a certain symmetry to the news: both are doubling down on their strategies, and it’s debatable which is taking the most risk by doing so.

Microsoft Buys LinkedIn

In its own small way perhaps the most surprising — and in many respects, encouraging — aspect of Microsoft’s purchase of LinkedIn was just how unexpected it was: a company with a notorious history of cutthroat boardroom politics not only pulled off the largest acquisition in its history without anyone knowing, it actually sat on the news regarding a signed letter of intent for a month! Granted, that meant limiting the information to an exceptionally small number of people, but there has been a lot of evidence over the past two years that Microsoft under CEO Satya Nadella has, at least at the highest levels of management, been more focused and aligned on where Microsoft needs to go than at any point under former CEO Steve Ballmer.

It was Steve Ballmer who led Microsoft’s previous largest acquisition, that of Nokia in 2013, a deal that from day one made no sense (and that was opposed by Nadella). It is that deal, though, that is perhaps the best place to start from when it comes to understanding this one.

I marveled last month at how adroitly Nadella had killed the Windows Phone business (or to put it more accurately, allowed Windows to figure out on their own that the platform had been dead from the beginning). The wind down of that business, though, and the ongoing shift of Windows to effectively “maintenance mode”, opens up room in Microsoft’s R&D budget — about 13% of revenue — for something new, so why not LinkedIn? Leaving aside the purchase price, LinkedIn slides right into the Windows Phone void when it comes to Microsoft’s investment in the future, with the benefit of being a business that has actual upside.

I do believe that upside is magnified significantly by Microsoft: should LinkedIn’s Sales Navigator, for example, sell in to 100% of the Microsoft Dynamics CRM user base, a good portion of this deal would be paid for; on the flipside, Dynamics now becomes a much more compelling offering in its ongoing competition with Salesforce (the most likely candidate for the rumored competitive bidder), Oracle, and SAP. And, of course, Microsoft’s Office products become that much more compelling with far deeper LinkedIn integration than was possible when they were two different companies (privacy concerns are much more easily solved when it’s the same company!).

I think, though, there is a deeper benefit that alters the trajectory of Microsoft’s productivity business in particular. I’ve written previously how Microsoft’s enterprise approach has been fundamentally upended by the cloud: when compatibility and ease-of-integration are no longer the controlling factors for IT buyers, Microsoft’s focus on lock-in and a good-enough user experience are simply not enough, and to the company’s credit both its Azure and Office 365 divisions have embraced a future that will be won on the user experience on all devices, not just Microsoft-controlled ones. Still, the question of who would own identity — previously the linchpin of Microsoft’s enterprise lock-in (thanks to Active Directory) — has been an open one.

What is potentially transformative about this deal is a future where Microsoft retains its focus on enterprise while shifting the locus of its business from companies to employees. I have written at length about the importance of owning the end user, but we the end users have multiple identities, one of which is our professional life, and that is a graph in which LinkedIn is nearly unchallenged. To put it another way, the “consumerization of IT” — a tagline favored by Ballmer — it not only about creating a compelling user experience in IT products but about treating enterprise users as, well, consumers: with LinkedIn Microsoft can form a direct relationship with its end users that goes far beyond the CIO and opens up a huge array of opportunities that not only were unavailable previously, but are also critical in a world where CIOs matter less than they ever have previously, and where employees change jobs constantly. Instead of starting from scratch with every new hire, it is Microsoft that is positioned to provide the glue that connects enterprise workers no matter where they are.

And, for what it’s worth, Microsoft got a deal: I think LinkedIn’s February stock slide was justified, but the fact remains that Microsoft is buying the social network for a price that would have been unthinkable six months ago, and the upside that justified the former stock price remains: LinkedIn knows more about its users than anyone outside of Facebook — and when it comes to our professional lives, they know more. This is the most valuable data in the world.

The most obvious criticism of this deal is the opportunity cost: what might that $26 billion have been spent on otherwise? Even with this deal Microsoft’s existential threat remains: why would a new company buy any of their products? This acquisition doesn’t solve that in a way a pairing of, say, Dropbox and Slack would (two tastes that would be better together, and who also aren’t selling), but then again, Microsoft has far more money than they do time: LinkedIn builds a bridge for their productivity business to a new world centered around end users, not corporations, and it may even give Microsoft’s inevitable compete products a head start. Imagine this: instead of simply moving Active Directory to the cloud, Microsoft is potentially making LinkedIn the central repository of identity for all business-based interactions: chat, email, and more, and it’s an identity that endures for an end user’s professional life, because it’s managed by the user, not by their transient employer. It’s genuinely exciting, and shocking though it may have been, I think that’s worth $26 billion.

More broadly, this has to be seen as the final blow to the notion that Windows is Microsoft’s focus; as late as 2014 I argued the company should split itself up, simply because the “Windows first” culture was so entrenched, but Nadella has done a remarkable job reorganizing and reorienting the company towards a future where Windows is one of many clients for Microsoft’s service offerings, and if it wasn’t clear to the rank-and-file yet, it surely must be now. Microsoft is doubling-down on the cloud and on productivity, and now there are 26 billion reasons to believe them.

Apple at WWDC

Apple’s WWDC announcements, meanwhile, were perhaps most notable for what they didn’t include, including no mention of iMessage on Android, and no significant discussion about Siri beyond the rumored and sorely needed incremental improvements (and even there, somewhat limited: Siri’s API is only open to a subset of applications).

Instead the keynote was about enhancing and deepening the value that comes from living the full Apple lifestyle: now your Watch unlocks your Mac, and your desktop is available on your iPhone. You can pay for things in your Mac’s browser using the iPhone, and control your house via the Apple TV from your lock screen. And yes, Messages got a massive update that I am very excited about (more tomorrow as I review the keynote in-depth), but its features are only available if you’re messaging other iPhone users: pay up or get left out.

Just as notable were the new things that weren’t announced: Apple is allowing users to delete pre-installed apps, and while there is no news about defining new default apps, should Apple do so the company would be on the path towards making its hardware mastery even more compelling by virtue of enabling services companies like Google and Microsoft to enhance the experience of using an iPhone.

This enhancement, of course, theoretically weakens Apple in the long run, as it squanders the ability of the company to leverage its hardware advantage into a services lock-in, but given my conviction on the power of culture I’m not sure this is a bad thing; here Steve Jobs’ advice rings true:

I think if you do something and it turns out pretty good, then you should go do something else wonderful, not dwell on it for too long. Just figure out what’s next.

For Apple what is next should almost certainly be guided by what the company is the best at: integrating hardware and software to deliver a user experience so compelling that consumers continue to self-select into the company’s own orbit, not building infrastructure on top of platforms it doesn’t control.

That, though, is also what made me nervous about the company’s announcements. CEO Tim Cook emphasized at the beginning of the keynote, in his words, “why we do what we do at Apple.”

Screen Shot 2016-06-14 at 6.04.37 AM

Our North Star has always been about improving people’s lives by creating great products that change the world.

Several of the product announcements, though, like enhancements to photos and Siri, seemed to care more about an absolutist view of privacy than about the best possible end user experience; make no mistake, I value privacy, but everything is a trade-off. At what point might Apple’s strenuous defense of privacy shift from a principled stand to a convenient reason to not be competitive with alternatives like Google photos or Alexa? Or, to put it another way, when does principle become an excuse for not being competitive on the user experience that has been Apple’s biggest differentiator for its entire existence?

Apple Versus Microsoft

Apple’s keynote was notable for its having doubled-down on what the company excels at: providing a better experience provided you pay Apple’s hardware margins. In this their announcement echoes Microsoft’s decision to double down on owning professional productivity and services. The difference, though, is in the timing: Microsoft is (finally) pivoting to the approach they should have adopted a decade ago, and while that has cost the company a lot of time and increased their risk, it is exciting to see them closer to the beginning than the end of their strategy. The question for Apple is where on that spectrum do they lie: for how long will a hardware-centric strategy drive growth, and just how much is Apple willing to change its culture to ensure it takes advantage of new opportunities and not simply preserve what it has?

Or maybe it doesn’t matter: odds are the biggest news from Monday will be Snapchat’s launch of an advertising API; the world goes on, value moves up the stack, and attention to the “story of the day” is all too often a trailing indicator.

Podcast: Exponent Episode 082 — A Podcast About Podcasts

Posted on

On the newest episode of Exponent, the podcast I co-host with James Allworth:

Ben and James discuss the present and future of podcasts, and how media companies need to take a holistic approach to their businesses.

Links

  • Ben Thompson: The Future of Podcasts — Stratechery
  • Ben Thompson: Why Web Pages Suck — Stratechery
  • Ben Thompson: Publishers and the Smiling Curve — Stratechery
  • Ben Thompson: Grantland and the (Surprising) Future of Publishing — Stratechery
  • Ezra Klein Show: Ben Thompson on How to Make it in the Media in 2016 — Vox
  • Marco Arment: Apple’s Actual Role in Podcasting: Be Careful What You Wish For — Marco.org
  • Podcasts Surge, but Producers Fear Apple Isn’t Listening — New York Times

Listen to the episode here

Podcast Information: Feed | iTunes | SoundCloud | Twitter | Feedback

The Future of Podcasting

Posted on

I like driving, even if I end up sitting in traffic. I enjoy doing the laundry, and take my time folding shirts just so. I volunteer to wash the dishes. After all, each of these activities is an excuse to listen to more podcasts.1

I’ve been listening to podcasts for over a decade now; I don’t remember exactly when I got started but it was around the time that Apple Took Podcasting Mainstream: that’s from the title of the press release announcing iTunes support for podcasts in 2005. Given that most podcasts were listened to on iPods (thus the name) that already synced with iTunes, Apple’s move dramatically simplified the distribution of podcasts: simply click a button in the music management app you already used, hook up the iPod as you already did, and voilà! New podcasts ready to be listened to in the car (via your cassette tape adaptor), while doing laundry, washing the dishes, etc. It was great!

It also was not in the slightest bit mainstream: according to Edison Research, in 2006 only 22% of Americans were even familiar with the term “podcasting”, and only 11% had ever listened to one. Both numbers have slowly but steadily grown over the years (55% have heard of podcasting as of this year, and 36% have listened to one, and there actually isn’t a readily apparent ‘Serial’ bump), aided in large part by the smartphone: by removing the need to sync with iTunes it was much easier to have fresh podcasts at the ready. Still, there remained the challenge of creating compelling content, discovering content worth listening to, retaining listeners and, of course, paying for it all.

Podcasting Versus Blogging

Late last year Joshua Benton wrote that Podcasting in 2015 Feels a Lot Like Blogging Circa 2004:

Podcasting is giving me a case of déjà vu…The variety and quality of work being done is thrilling; outside attention is growing; new formats are evolving. We’re seeing the same unlocking of creative potential we saw with blogging, and there’s far more good work being produced than anyone has time to take in. The question now is whether podcasting’s future will play out as the last decade of blogging has.

It’s a good observation, but there are important differences if you look into the various factors I alluded to above:

  • Creation: Blogger was released back in 1999, and WordPress in 2003; both required some level of acumen, but significantly less than recording and mixing a podcast takes today. It’s also very complicated to get a show actually listed in iTunes. This means, by extension, that for all the great podcasts there are today, there were that many more blogs.
  • Distribution: Blogs could be read via URLs typed in a browser that everyone already used. Podcasts are much more complicated: you either have to search a 3rd-party podcast player’s directory (iTunes or self-contained) to add a show, or copy-and-paste a feed address. Alternately, you can simply listen on a website, but that is a suboptimal experience to say the least.
  • Discovery: In 2004 most blogs were found from links on more popular blogs; today new blogs are usually discovered on social networks. Podcasts, meanwhile, really struggle here: yes, iTunes has a front page and a blackbox ranking system, but the requirement to download a file and spend time listening makes it hard to spread virally. Many podcasts are instead built off of established brands like NPR or the personal brands of the podcast hosts.
  • Retention: Back in 2004 most blog readers returned via bookmarks; more advanced users leveraged RSS readers that polled sites for new content and downloaded it in a feed. Today, most readers rely on social network posts that may or may not be seen. Interestingly, it is here that podcasts have an advantage: because they are built on RSS, anyone who “subscribes” through a podcast player downloads podcasts automatically and even gets notifications, making for a very sticky and loyal audience.
  • Monetization: Blogging had a brief honeymoon period where you could actually make money with Google AdSense, but revenue soon plummeted as inventory drastically increased; more devastating, not just for bloggers but all publishers, was Facebook’s absorption of not just what used to be blogging content but also publishing dollars. Increasingly, the best option for publishers is to simply publish directly to Facebook and let them sell ads.

The monetization of podcasts, meanwhile, deserves a deeper dive.

The Problem with Monetizing Podcasts

Podcasting is still a tiny business: according to the Wall Street Journal podcasts only attracted $34 million in advertising revenue last year, about 1/100th the amount spend on billboards (This number is disputed by many of the bigger podcasters, but the highest estimate I’ve seen is $200 million, i.e. 1/15th of billboards). The biggest player in the podcast advertising market is a company called Midroll Media, which was acquired by E.W. Scripps last year for $50 million (with a $10 million earn-out).

Midroll sells ads for over 200 podcasts, including some of the most popular ones like WTF with Marc Maron and the Bill Simmons Podcast. The not-so-secret reality about podcast ads, though, are that advertisers are quite concentrated: a FiveThirtyEight intern heroically listened to the top 100 shows on the iTunes chart and counted 186 ads; 35 percent of them were from five companies.2 More tellingly, nearly all of the ads were of the direct marketing variety.

A major challenge in podcast monetization is the complete lack of data: listeners still download MP3s and that’s the end of it; podcasters can measure downloads, but have no idea if the episode is actually listened to, for how long, or whether or not the ads are skipped. In a complete reversal from the online world of text, the measurement system is a big step backwards from what came before: both radio and TV have an established measurement system for what shows are watched, and the scale of advertising is such that surveys can measure advertising effectiveness. Thus the direct marketing advertisers: they can simply do the measurement themselves through coupon codes or special URLs that measure how many people responded to a podcast ad. It’s not totally efficient — some number of conversions forget the code or URL — but it’s something.

It also won’t scale. For the advertisers that exist the implication of measuring by code or URL is that every single podcast needs customized support, limiting advertising opportunities to bigger podcasts only. More importantly, there simply aren’t that many advertisers with the sort of business model that can justify the hassle. The real money in TV and especially radio is brand advertising; brand advertising is focused on building affinity for a purchase that will happen at some indefinite point in the future, so the focus is less on conversion and more on targeting: knowing in broad strokes who is listening to an ad, and exactly how many people. For podcasting to ever be a true moneymaker it has to tap into that — and that means changing the fundamental nature of the product.

Midroll Makes Their Move

Yesterday Scripps/Midroll made another acquisition, this time of the podcast player Stitcher. From the Wall Street Journal:

Stitcher is a free app that streams more than 65,000 podcasts from publishers ranging from NPR to MSNBC to The Wall Street Journal. It will operate under Midroll Media, the podcast advertising company that Scripps acquired last year…

“We certainly have the ad sales force and the connections that make us a leader in the space, but today we depend almost exclusively on distribution into other channels,” said Adam Symson, chief digital officer at Scripps. “This puts in place, with a very strong brand, another piece of the puzzle in the ecosystem play.”

As new listeners and shows enter the podcast world, companies in the space have been contending with a handful of industry challenges, like measuring audience size and wooing big brand marketers. “For the first time we’ll have significantly more ability to help with podcast discovery, to help with distribution, to help shows grow, and to help find out what audiences want in a way that we could not do before,” Mr. Diehn said.

Stitcher is thought to be the 2nd most popular podcast player, although it has long been controversial in some circles for its default practice of hosting podcasts itself (instead of directing users to download them directly from a podcaster’s server) and inserting ads. That model, though, was likely attractive to Scripps/Midroll: controlling the files and the player means the possibility of making meaningful measurements of play data plus dynamic ad insertion at scale.

Moreover, Midroll’s leading role in advertising combined with Scripps’ bank account mean the company could offer big bucks to leading podcasters to make themselves exclusive to Stitcher, driving users to the measurable app to the long-term benefit of the company’s efforts to attract brand advertisers. To be very clear, there are a lot of obstacles to this actually happening, but the idea of there being a central aggregator for podcasts that locks in podcasters through superior monetization and listeners through exclusive content is both plausible and attractive from a business perspective.

Publishers Beware

Popular podcaster and blogger Marco Arment has been particularly vocal about the problems with this approach; more pertinently, Arment says he built the Overcast podcasting app to resist this outcome:

Podcasts are hot right now. Big Money is coming. Big Money isn’t going to sell nicely designed, hand-crafted, RSS-backed podcast players for $2.99 or ask you to pay what you want to support them, because that doesn’t make Big Money. They’re coming with shitty apps and fantastic business deals to dominate the market, lock down this open medium into proprietary “technology”, and build empires of middlemen to control distribution and take a cut of everyone’s revenue…I don’t know if Overcast stands a chance of preventing the Facebookization of podcasting, but I know I’m increasing the odds if my app is free without restrictions. As long as I can make money some other way, I’m fine.

That phrase — “the Facebookization of podcasting” — should send chills down the spine of all the publishing companies jumping headfirst into podcasting. Publishers are already “serfs in a kingdom that Facebook owns” by virtue of the fact that Facebook owns user attention and has superior advertising capabilities. And yet many publishers are so focused on finding new income streams that they are practically begging for exactly what Arment fears.

Early last month the New York Times suggested many publishers hoped the Facebook of podcasting would be Apple; thanks to that 2005 release the company and its directory remain the center of the podcasting industry, and iTunes and its iOS podcast app are are responsible for a reported 65 percent share of podcast listeners. Indeed, this is the biggest reason to doubt Midroll’s plans: as difficult as it can be to corral advertisers, switching the habits of millions of listeners in the face of a default experience is far more difficult.

A Third Way

All that said, I’m not sure the status quo of podcasters hosting their own MP3 files listed in a (relatively) open directory mostly ignored by Apple is sustainable, or even desirable: relatively large independent podcasters like Arment may prefer the current setup, but there is an increasing amount of money and agitation for building something that looks a lot like Midroll’s presumed plans for Stitcher. Apple itself, with its dominant position in podcasting and its newfound focus on services revenue in the face of declining iPhone sales, is not only well-placed but also increasingly motivated to fill that role itself.

More importantly, though, for publishers podcasting really is a great opportunity to build something sustainable. In Grantland and the (Surprising) Future of Publishing I explained how media companies need to expand their thinking about monetization:

Too much of the debate about monetization and the future of publishing in particular has artificially restricted itself to monetizing text. That constraint made sense in a physical world: a business that invested heavily in printing presses and delivery trucks didn’t really have a choice but to stick the product and the business model together…

Focused, quality-obsessed publications [should]…collect “stars” and monetize them through…alternate media forms. Said media forms, like podcasts, are tough to grow on their own, but again, that is what makes them such a great match for writing, which is perfect for growth but terrible for monetization.

Go back to the five factors that go into effective media: both text and podcasts are relatively easy to create, but text is much easier to distribute and discover; the most effective podcasts, meanwhile, are those driven by brands or personalities; podcasts in general are great at retaining loyal listeners; and their monetization potential is much higher if the measurement can be figured out.

A Stitcher/Apple-type solution does help on that last point, but it still makes distribution and discovery harder than they should be: a publisher has to tell its readers to go to a different app, search for their name, subscribe, and then depend on that 3rd party app for monetization and measurement. Wouldn’t it be better if the publisher simply did that themselves?

I think there is a third way here, that preserves independence but starts to solve the monetization and measurement problem: publishers should offer podcasts through their own app that measures listens, and either sell ads themselves if they have the scale or outsource it to a company like Midroll.3 Midroll, for their part, should leverage their new player technology to offer skinnable apps for publishers who can’t build their own. The end result would be a much smoother path for publishers to convert their readers to listeners — and to effectively cross-promote — along with the measurement and scale needed to grow advertising meaningfully (or even offer subscriptions).4

I know this breaks the modern concept of podcasting, and power users with tens of subscriptions in their podcasting player of choice will be annoyed if they have to download multiple apps. Often, though, a solution that works for power users is actually prohibitive for normal users, and the other solution — a Facebook of podcasts — would be worse for everyone. Just look at what happened to RSS readers: yes, Google killed them, but it was only ever used by a fraction of readers before then because they were too difficult; Facebook, on the other hand, was easy.5 Fortunately for publishers, the challenges of podcast discovery and distribution actually make apps the easiest choice of all.6

There are plenty of good reasons why the publishing world ended up subservient to Facebook, but to answer Benton’s question as to “whether podcasting’s future will play out as the last decade of blogging has”, I don’t think it has to. The limitations of audio relative to text actually work to the publishers’ advantage in a way that the portability of text did not, and this may be their last chance to build destinations that people will wash the dishes in order to visit.

  1. I mostly listen to sports podcasts, primarily NBA, so no need to email for my list of recommendations 🙂  ↩

  2. Squarespace had 30, Stamps.com 12, Audible 11, MailChimp 8, and Dollar Shave Club 5 ↩

  3. I’m using Midroll as a standin, but this role could be filled by any company ↩

  4. Also, there’s no need for publishers to fear the App Store: Apple doesn’t take a skim off of advertising, and the App Store infrastructure would actually make subscriptions viable ↩

  5. To be clear, I love RSS! And it underpins a lot of the web, including Facebook Instant Articles. But I’m talking about the mass market and monetization ↩

  6. Yes Newsstand was a failure; however, the entire premise of this article is that text is different than audio ↩

Podcast: Exponent Episode 081 — We Have Always Been at War with Amazon

Posted on

On the newest episode of Exponent, the podcast I co-host with James Allworth:

Ben and James explain why Alexa is an even bigger deal than it seems, and why both Apple and Google should fear Amazon. Plus, why Microsoft’s challenges are only beginning.

Links

Listen to the episode here

Podcast Information: Feed | iTunes | SoundCloud | Twitter | Feedback

Exponent Podcast: Multiple Full Circles

Posted on

On Exponent, the weekly podcast I host with James Allworth, we revisit last week’s discussion on Facebook, then discuss the Peter Thiel versus Gawker affair, and the implications for tech broadly. Then, how culture helps and hurts at Facebook, Apple, Google, and Microsoft.

Listen to it here.

The Curse of Culture

Posted on

One of the seminal books on culture is Edgar Schein’s Organizational Culture and Leadership. Schein writes in the introduction:

Perhaps the most intriguing aspect of culture as a concept is that it points us to phenomena that are below the surface, that are powerful in their impact but invisible and to a considerable degree unconscious. In that sense, culture is to a group what personality or character is to an individual. We can see the behavior that results, but often we cannot see the forces underneath that cause certain kinds of behavior. Yet, just as our personality and character guide and constrain our behavior, so does culture guide and constrain the behavior of members of a group through the shared norms that are held in that group.

In Schein’s telling, things like ping pong tables and kegerators are two (small) examples of artifacts — the visible qualities of an organization. They are easy to observe but their meaning is usually indecipherable and unique to a particular group (to put it another way, copying Google’s perks is missing the point).

The next level down are espoused beliefs and values, what everyone in an organization understands consciously: “openness,” for example, or “the customer is always right”; as you might expect espoused beliefs and values devolve rather easily into cliché.

It’s the third level that truly matters: underlying assumptions. Schein writes:

Basic assumptions, in the sense in which I want to define that concept, have become so taken for granted that one finds little variation within a social unit. This degree of consensus results from repeated success in implementing certain beliefs and values, as previously described. In fact, if a basic assumption comes to be strongly held in a group, members will find behavior based on any other premise inconceivable.

The implications of this definition are profound: culture is not something that begets success, rather, it is a product of it. All companies start with the espoused beliefs and values of their founder(s), but until those beliefs and values are proven correct and successful they are open to debate and change. If, though, they lead to real sustained success, then those values and beliefs slip from the conscious to the unconscious, and it is this transformation that allows companies to maintain the “secret sauce” that drove their initial success even as they scale. The founder no longer needs to espouse his or her beliefs and values to the 10,000th employee; every single person already in the company will do just that, in every decision they make, big or small.

Microsoft’s Blindness

As with most such things, culture is one of a company’s most powerful assets right until it isn’t: the same underlying assumptions that permit an organization to scale massively constrain the ability of that same organization to change direction. More distressingly, culture prevents organizations from even knowing they need to do so. Schein continues:

Basic assumptions, like theories-in-use, tend to be nonconfrontable and nondebatable, and hence are extremely difficult to change. To learn something new in this realm requires us to resurrect, reexamine, and possibly change some of the more stable portions of our cognitive structure…Such learning is intrinsically difficult because the reexamination of basic assumptions temporarily destabilizes our cognitive and interpersonal world, releasing large quantities of basic anxiety. Rather than tolerating such anxiety levels, we tend to want to perceive the events around us as congruent with our assumptions, even if that means distorting, denying, projecting, or in other ways falsifying to ourselves what may be going on around us. It is in this psychological process that culture has its ultimate power.

Probably the canonical example of this mindset was Microsoft after the launch of the iPhone. It’s hard to remember now, but no company today comes close to matching the stranglehold Microsoft had on the computing industry from 1985 to 2005 or so.1 The company had audacious goals — “A computer on every desk and in every home, running Microsoft software” — which it accomplished and then surpassed: the company owned enterprise back offices as well. This unprecedented success changed that goal — originally an espoused belief — into an unquestioned assumption that of course all computers should be Microsoft-powered. Given this, the real shock would have been then-CEO Steve Ballmer not laughing at the iPhone.

A year-and-a-half later, Microsoft realized that Windows Mobile, their current phone OS, was not competitive with the iPhone and work began on what became Windows Phone. Still, unacknowledged cultural assumptions remained: one, that Microsoft had the time to bring to bear its unmatched resources to make something that might be worse at the beginning but inevitably superior over time, and two, that the company could leverage Windows’ dominance and their Office business. Both assumptions had become cemented in Microsoft’s victory in the browser wars and their slow-motion takeover of corporate data centers; in truth, though, Microsofts’ mobile efforts were already doomed, and nearly everyone realized it before Windows Phone even launched with a funeral for the iPhone.

Steve Ballmer never figured it out; his last acts were to reorganize the company around a “One Microsoft” strategy centered on Windows, and to buy Nokia to prop up Windows Phone. It fell to Satya Nadella, his successor, to change the culture, and it’s why the fact his first public event was to announce Office for iPad was so critical. I wrote at the time:

This is the power CEOs have. They cannot do all the work, and they cannot impact industry trends beyond their control. But they can choose whether or not to accept reality, and in so doing, impact the worldview of all those they lead.

Microsoft under Nadella’s leadership has, over the last three years, undergone a tremendous transformation, embracing its destiny as a device-agnostic service provider; still, it is fighting the headwinds of Amazon’s cloud, open source tooling, and the fact that mobile users had six years to get used to a world without Microsoft software. How much stronger might the company have been had it faced reality in 2007, but the culture made that impossible.

Steve Jobs’ Leadership

Shein defines leadership in the context of culture:

When we examine culture and leadership closely, we see that they are two sides of the same coin; neither can really be understood by itself. On the one hand, cultural norms define how a given nation or organizations will define leadership—who will get promoted, who will get the attention of followers. On the other hand, it can be argued that the only thing of real importance that leaders do is to create and manage culture; that the unique talent of leaders is their ability to understand and work with culture; and that it is an ultimate act of leadership to destroy culture when it is viewed as dysfunctional.

A great example of this sort of destruction was Steve Jobs’ first keynote as interim CEO at the 1997 Boston Macworld, specifically the announcement of Apple’s shocking partnership with Microsoft:

When Jobs said the word Microsoft, the audience audibly groaned. A few minutes later, when Jobs clicked to a slide that said Internet Explorer would be the default browser on Macintosh, the audience booed so loudly that Jobs had to stop speaking. When Jobs finally said the actual words “default browser” the audience booed even louder, with several individuals shouting “No!” It is, given the context of today’s Apple keynotes, shocking to watch.

Then, after Bill Gates spoke to the crowd via satellite (in what Jobs would call his “worst and stupidest staging event ever”), Jobs launched into what his biographer Walter Isaacson called an “impromptu sermon”:

If we want to move forward and see Apple healthy and prospering again, we have to let go of a few things here. We have to let go of this notion that for Apple to win Microsoft has to lose. OK? We have to embrace a notion that for Apple to win Apple has to do a really good job, and if others are going to help us, that’s great, cause we need all the help we can get. And if we screw up and we don’t do a good job, it’s not somebody else’s fault. It’s our fault. So, I think that’s a very important perspective.

I think, if we want Microsoft Office on the Mac, we better treat the company that puts it out with a little bit of gratitude. We like their software. So, the era of setting this up as a competition between Apple and Microsoft is over as far as I’m concerned. This is about getting healthy, and this is about Apple being able to make incredibly great contributions to the industry, to get healthy and prosper again.

Here’s Shein:

But as the group runs into adaptive difficulties, as its environment changes to the point where some of its assumptions are no longer valid, leadership comes into play once more. Leadership is now the ability to step outside the culture that created the leader and to start evolutionary change processes that are more adaptive. This ability to perceive the limitations of one’s own culture and to evolve the culture adaptively is the essence and ultimate challenge of leadership.

Make no mistake: even though he had been gone for over a decade, Steve Jobs was responsible for that booing.

jobsibm121230-1

Jobs had set up Apple generally and the Macintosh specifically as completely unique and superior to the alternatives, particularly the hated IBM PC and its Windows (originally DOS) operating system. By 1997, though, Microsoft had won, and Apple was fighting for its life. And yet the audience booed its lifeline! That is how powerful culture can be — and that is why Jobs’ “impromptu sermon” was so necessary and so powerful. It was Apple’s version of Office on the iPad, and a brilliant display of leadership.

Warning Signs for Apple and Google

Over the weekend Marco Arment wrote a widely-read piece (now) called If Google’s Right About AI, That’s a Problem for Apple:

The BlackBerry’s success came to an end not because RIM started releasing worse smartphones, but because the new job of the smartphone shifted almost entirely outside of their capabilities, and it was too late to catch up. RIM hadn’t spent years building a world-class operating system, or a staff full of great designers, or expertise in mass production of luxury-quality consumer electronics, or amazing APIs and developer tools, or an app store with millions of users with credit cards already on file, or all of the other major assets that Apple had developed over a decade (or longer) that enabled the iPhone. No new initiative, management change, or acquisition in 2007 could’ve saved the BlackBerry. It was too late, and the gulf was too wide.

Today, Amazon, Facebook, and Google are placing large bets on advanced AI, ubiquitous assistants, and voice interfaces, hoping that these will become the next thing that our devices are for. If they’re right — and that’s a big “if” — I’m worried for Apple…If the landscape shifts to prioritize those big-data AI services, Apple will find itself in a similar position as BlackBerry did almost a decade ago: what they’re able to do, despite being very good at it, won’t be enough anymore, and they won’t be able to catch up.

Arment is exactly right. What is fascinating, though, is that, as I wrote last week, Google has their own set of problems: users actually spend their time in social apps, mostly owned by Facebook, and while Google has a critical asset in Android, its most valuable users (from a monetization standpoint) are on iOS. How will users actually access Google’s AI capabilities (if they turn out to matter), and how will Google monetize them?

To be sure, neither company is struggling today. Apple may have failed to achieve record results for the first time in 13 years, but their 2Q 2016 revenue of $50.6 billion was more than the revenue of Microsoft, Google, and Facebook combined; Google, meanwhile, is still setting year-over-year records, with $17.3 billion in revenue.

That, though, is the challenge: BlackBerry wasn’t struggling in 2006, nor was Microsoft in 2007, or even Apple as late as 1993. There was no obvious reason to think that anything was amiss, and it was culture that ensured that whatever hints there were would be ignored. Shein again:

Culture as a set of basic assumptions defines for us what to pay attention to, what things mean, how to react emotionally to what is going on, and what actions to take in various kinds of situations. Once we have developed an integrated set of such assumptions—a “thought world” or “mental map”—we will be maximally comfortable with others who share the same set of assumptions and very uncomfortable and vulnerable in situations where different assumptions operate, because either we will not understand what is going on, or, worse, we will misperceive and misinterpret the actions of others.

And so BlackBerry thought Apple was lying about the iPhone; Steve Ballmer declared “He liked Microsoft’s chances”; and Apple, well, Apple had already decided to, in Jobs’ view, sacrifice product for profits. The time to act was at the moment of denial, not the moment of crisis.

Paths Forward

That said, both Apple and Google are still operating from positions of considerable strength going forward: iPhone growth may or may not have peaked, but it’s not going anywhere for a good long while, and the company is almost certainly working on a car. Google, meanwhile, is arguably in even better shape: the company has a massive lead in machine learning, which could manifest itself in all kinds of interesting applications, and here Android looms large.

Still, there are very obvious steps both companies could do to entrench their advantages:

  • Apple could partner with a company like Microsoft (again) to build out its services layer, both on the backend (Azure) and, if they want to get really radical, the front-end (combining Siri and Cortana). The most radical solution, though, would be fully opening up iOS in such a way that users could set Google (or any other company’s) services as defaults. This would foreclose any medium-term threat to the iPhone from an Android experience that is fully-infused with Google’s AI capabilities (more on the long-term problems in a moment)
  • Google could — should! — build a bot for Facebook Messenger. More than that, they should build an entire backend for Facebook Messenger developers. Do people want to live in Facebook? Very well, meet them there, just as Google found its user base on Windows through the browser.

Both ideas (and there are certainly others) have their issues: Apple would be foreclosing their future as a services provider, but frankly, I am extremely skeptical about this regardless. Not only does the company have the wrong organizational structure but, similar to Microsoft, the company’s overwhelming success has had far-reaching effects on the culture; in this case, the company is so focused on making physical products that it’s doubtful an effective services mentality could ever emerge, not to mention the company’s (at times disingenuous) absolutism about privacy.2

Google, meanwhile, would be supporting its most dangerous competitor. At the end of the day Google and Facebook share the exact same customers — advertisers — and even though it’s not clear how Google can steal attention back it’s also not obvious that they should aid their rival.3

The Curse of Culture

The biggest problem for both, though, is culture. Apple, beyond everything else — and in part because of the humiliation of that 1997 keynote — desires complete control; Google, for its part, desires information, and can’t tolerate the idea of Facebook having more.

The rigidity of both is the manifestation of the disease that affects every great company: the assurance that what worked before will work eternally into the future, even if circumstances have changed. What makes companies great is inevitably what makes companies fail, whenever that day comes.4

  1. Yes, Apple ultimately came to earn much more revenue that Microsoft ever did, and Google has come close, but both did so in the context of a much larger industry ↩

  2. This too is why I don’t buy the “Wait for WWDC” response to Marco’s article; the reasons to be skeptical about Apple’s prospects here are structural ↩

  3. That, in some respects, gets to the tragedy of this piece: Apple and Google are the most natural of partners. Neither has to lose for the other to win, and both have wasted far too much valuable time fighting a war that was never necessary. ↩

  4. One final quote from Shein:

    If one wishes to distinguish leadership from management or administration, one can argue that leadership creates and changes cultures, while management and administration act within a culture. By defining leadership in this manner, I am not implying that culture is easy to create or change, or that formal leaders are the only determiners of culture. On the contrary, as we will see, culture refers to those elements of a group or organization that are most stable and least malleable. Culture is the result of a complex group learning process that is only partially influenced by leader behavior. But if the group’s survival is threatened because elements of its culture have become maladapted, it is ultimately the function of leadership at all levels of the organization to recognize and do something about this situation. It is in this sense that leadership and culture are conceptually intertwined.

    Are Tim Cook and Sundar Pichai managers, or leaders? And which do they need to be? ↩

Google’s Go-to-Market Gap

Posted on

Perhaps the most surprising aspect of Google’s rise is that it is almost entirely attributable to having the best technology. That sounds like it should be the normal state of affairs, but in truth there are an untold number of research projects and startups that had superior technology but never became viable businesses; perhaps there was no business model, or an inability to build a requisite ecosystem, or most commonly, an inability to find a viable market and/or reach consumers who might be interested.

Great Companies Versus Great Technology

For example, look at the other technology giants, all of whom got their start on the basis of more than pure technology:

  • While Bill Gates and Paul Allen built Microsoft’s first product (Altair BASIC), the company’s dominance was established via a business development deal with IBM to provide an operating system for the nascent IBM personal computer; the actual OS (MS-DOS) was acquired from a company called Seattle Computer Products. And while Microsoft would go on to develop all kinds of technology, everything that followed rested on the leverage from that IBM deal.
  • Amazon started out as a primitive website that was differentiated by its selection and ability to deliver anywhere in the U.S. And while the company has certainly invented a lot of technology when it comes to web services and logistics, its advantage remains rooted in its scale.
  • Facebook’s technology was so basic that Mark Zuckerberg’s first employee — his roommate Dustin Moskovitz — didn’t even know how to program; he would go on to be Facebook’s first Chief Technical Officer. What got the site off the ground was the way it digitized pre-existing offline networks — it started from its market and worked backwards.
  • Apple’s strategy has certainly been predicated on having the best products, but that does not necessarily mean the company has always had the best technology. The Mac GUI was famously “inspired” by Xerox PARC, the iPod was hardly the first MP3 player, and while the original iPhone was certainly a technological marvel, it not only was built on everything that came before it but also required huge investments in distribution to become the juggernaut it is1

To be clear, all of these companies had great technology, but it wasn’t enough — it rarely is.

Google = Best

Google stands in stark contrast: relying on links and a lot of math to rank sites was a technological breakthrough of the first order — and no company wanted to buy it, despite the fact it was very much on sale. And yet, usage grew exponentially thanks to word-of-mouth: Google’s search was so startlingly better — and the cost of trying it was simply typing in a URL — that the product grew like wild fire without business development, distribution, or marketing. By the time Google did their first distribution deal, with Yahoo in 2000, Google was already handling millions of queries a day simply because they were superior; Yahoo only hastened Google’s inevitable domination.

The focus on being the best became a core piece of Google’s identity, and the biggest factor in how they hired. Steven Levy wrote in In the Plex:

The founders also knew that Google had to be a lot smarter to keep satisfying users—and to fulfill the world-changing ambitions of its founders. “We don’t always produce what people want,” Page explained in Google’s early days. “It’s really difficult. To do that you have to be smart—you have to understand everything in the world. In computer science, we call that artificial intelligence.”

Brin chimed in. “We want Google to be as smart as you—you should be getting an answer the minute you think of it.”

“The ultimate search engine,” said Page. “We’re a long way from that.”

Page and Brin both held a core belief that the success of their company would hinge on having world-class engineers and scientists committed to their ambitious vision. Page believed that technology companies can thrive only by “an understanding of engineering at the highest level”…

“We just hired people like us,” says Page.

So many of Google’s successes — and failures — is wrapped up in this sentiment. So, too, is their future.

The Google Assistant

Yesterday at the Google I/O keynote the dominant theme was the very real progress Google is making on genuine Artificial Intelligence that goes far beyond search. Sundar Pichai said in his opening remarks:

It’s amazing to see how people engage differently with Google. It’s not just enough to give them links. We really need to help them get things done in the real world. This is why we are evolving search to be much more assistive. We’ve been laying the foundation for this for many, many years through investments in deep areas of computer science. We’ve built the knowledge graph — we today have an understanding of 1 billion entities, people, places, and things, and the relationships between them and the real world. We have dramatically improved the quality of our voice recognition…Image recognition and computer vision, we can do things we never thought we could do before…We even do real-time translation.

Progress in all of these areas is accelerating, thanks to profound advances in machine learning and [artificial intelligence] (AI), and I believe we are at a seminal moment. We as Google have evolved significantly over the past ten years and we believe we are poised to take a big leap forward in the next ten years leveraging out state-of-the-art capabilities in machine learning and AI, we truly want to take the next step in being more assistive for our users. So today, we are announcing the Google assistant.

There is little question that Google is far ahead in artificial intelligence. Late January, in a humorous juxtaposition that was almost certainly coincidental but telling all the same, Facebook CEO Mark Zuckerberg posted about the social network company’s progress in building a computer that could play the board game ‘Go’, long thought unbeatable by computers. Mere hours later Demis Hassabis, the head of Google’s DeepMind division, revealed in a blog post that Google had done exactly that: their machine learning-based program, called AlphaGo, had defeated a three-time European champion, and would soon take on the best Go player in the world (AlphaGo would go on to win that match 4–1).

To be sure, this is a single example, but any time spent using the increasing number of Google products that rely on machine learning-based artificial intelligence — translation, voice and image recognition, and yes, search — quickly make it obvious just how much better Google is, and, thanks to the copious amount of data at the company’s disposal, how much better they are likely to become. The problem is that in today’s world being the best may not be enough.

Open Versus Closed

While describing how Google search grew by word-of-mouth, I snuck in one line that looms very large when it comes to thinking about both Google’s past and its future: “the cost of trying it was simply typing in a URL.” Google’s initial success was not just because they were superior at search: thanks to the fact that the interface with Google was a web page, the company had instant access to every person on earth with a PC and a functioning Internet connect — and they didn’t have to pay a dime. On the flip-side, if you heard about this amazing new search engine, you didn’t need to go buy a CD or even download a program: you simply typed “Google.com” and the results spoke for themselves. Make no mistake: the brilliance of Larry Page and Sergey Brin was only perhaps surpassed by the brilliance of the people they hired, particularly in the early days, but the company’s success was very much intertwined with the openness afforded by a browser and the world wide web.

Today, though, the PC is fading in relevance, and the browser along with it: what matters is mobile, and the means to connect with users is to either be embedded into the phone or have an app where people live. And while Google has a massive foothold thanks to Android, a huge number of its best customers are on iOS, and nearly all its customers live in Facebook.

The implications of this are obvious — just look at maps. Google Maps is widely regarded as being the superior product to Apple Maps, yet the latter is used three times as often on iPhones; such is the power of defaults and being “good enough.”2 Similarly, while Google’s voice recognition far outpaces Apple’s Siri, the fact that Apple sets the rules means that Google’s Gboard keyboard for iOS cannot include dictation.3 More broadly, on iOS the only way to use the Google assistant that Pichai announced yesterday will be to open a Google app (or go to a search field in, you guessed it, a browser): using Siri will always be much easier and frictionless.

The situation is even more challenging when it comes to social networks broadly and messaging specifically, which is to mobile as the browser was to the PC: a meta-OS where people spend the vast majority of their time. The problem for Google is that while the browser was an open platform that not even Microsoft could control — sure, they killed Netscape, but Google built its audience from within Internet Explorer — social networks and messaging services are not only closed but nearly impossible to compete with. No matter how great of a messaging service Google may build — another I/O announcement was a messaging service called Allo, which heavily features the Google assistant — the most important feature of any messaging service is whether or not your friends use it, and nearly every geography in the world is locked up by a competitor.

There is a new arena — the home, the one place where talking is usually better than pecking away at a phone no longer in your pocket — but here Google is behind Amazon. The latter, thanks to the failure of its own smartphone efforts, was freed from the smartphone obsession that resulted in Google wrongly identifying the smartphone-dependent Nest as its connected home offering, instead of a voice-focused standalone device like the Echo. There is almost certainly time to catchup, but it’s telling that Google’s announced competitor — Google Home — is still months away.

Google’s Go-to-Market Challenge

The net result is that Google has no choice but to put its founding proposition to the ultimate test: is it enough to be the best? Can the best artificial intelligence overcome the friction that will be involved in using Google assistant on an iPhone? Can the best artificial intelligence actually shift human networks? Can the best artificial intelligence win the home in the face of a big head start?

That the answer may very well be “no” (or mixed, at best), is at the root of my 2014 piece Peak Google. That piece was about business relevance, something that goes beyond the collection of cash or the creation of superior technologies. The question I was asking was which companies are the best equipped to build new businesses going forward, and here Google’s outlook is far cloudier than it was back when the company was, for all intents and purposes, invented.

The problem is that as much as Google may be ahead, the company is also on the clock: every interaction with Siri, every signal sent to Facebook, every command answered by Alexa, is one that is not only not captured by Google but also one that is captured by its competitors. Yes, it is likely Apple, Facebook, and Amazon are all behind Google when it comes to machine learning and artificial intelligence — hugely so, in many cases — but it is not a fair fight. Google’s competitors, by virtue of owning the customer, need only be good enough, and they will get better. Google has a far higher bar to clear — it is asking users and in some cases their networks to not only change their behavior but willingly introduce more friction into their lives — and its technology will have to be special indeed to replicate the company’s original success as a business.4

  1. To put it another way, the technology at the heart of Apple’s products — OS X and iOS — has its roots in NeXT, a business failure ↩

  2. By most accounts Apple Maps is indeed “good enough” in the U.S.; from personal experience, though, it very much falls short in many other countries ↩

  3. Google does deserve a lot of credit for finally remembering that Android exists to serve Google, which should be focused on all users, not its own platforms ↩

  4. Which itself is under threat: to fully leverage Google assistant in Google search will almost certainly deepen Google’s antitrust troubles with the European Union ↩

The Real Problem With Facebook and the News

Posted on

I got my start writing for the student newspaper at the University of Wisconsin.1

What is interesting about that statement is that the appropriate follow-up question is “Which student newspaper?” For many years Wisconsin was unique in being the only university with two daily newspapers, both with five-digit print circulations.2 The older paper, The Daily Cardinal, got its start in 1892, but in 1969, as Wisconsin became ground zero for some of the most intense protests against the Vietnam War, a group of conservative students, with support from right-wing luminary William F. Buckley, resolved to counter what they saw as a pervasive liberal bias from The Daily Cardinal specifically and media generally.

Against all odds the fledgling paper survived — and it’s those odds that interest me most. To start a paper in 1969 required a not insignificant amount of money to pay for everything from desks to typewriters to, most pertinently, (renting time on) a printing press. The reality is that Wisconsin was a huge aberration, not only amongst universities but amongst cities generally: most had one paper, maybe two, and there were only three broadcast TV networks.

This was an arrangement that was certainly profitable for those who owned these geographic monopolies, but it also had a curious effect on how news was experienced in the United States: first, there was a strict wall built between the editorial and business sides of a business (a wall that hinders publishers today), and secondly, befitting their dominant market position (and, perhaps, in a careful attempt to ensure they kept it), news organizations adopted a “balanced” he-said/she-said approach to reporting that Jay Rosen has characterized as The View From Nowhere.

The problem with this approach is that no matter how scrupulous a reporter or editor may be, they are still human, constrained to a world view informed by their own limited experiences, and, as was so often the case in nearly every professional workplace in America, those experiences were shared: white, middle to upper class, often from the coasts, educated at elite universities. And so began a longstanding conservative critique of the media: that while it claims to be balanced, what was actually printed or broadcast, both in terms of selection and tone, had a liberal bias.3

Facebook Trending News

Yesterday Gizmodo published a bombshell where the headline basically says it all: Former Facebook Workers: We Routinely Suppressed Conservative News.

Facebook workers routinely suppressed news stories of interest to conservative readers from the social network’s influential “trending” news section, according to a former journalist who worked on the project. This individual says that workers prevented stories about the right-wing CPAC gathering, Mitt Romney, Rand Paul, and other conservative topics from appearing in the highly-influential section, even though they were organically trending among the site’s users.

Several former Facebook “news curators,” as they were known internally, also told Gizmodo that they were instructed to artificially “inject” selected stories into the trending news module, even if they weren’t popular enough to warrant inclusion—or in some cases weren’t trending at all. The former curators, all of whom worked as contractors, also said they were directed not to include news about Facebook itself in the trending module.

In other words, Facebook’s news section operates like a traditional newsroom, reflecting the biases of its workers and the institutional imperatives of the corporation. Imposing human editorial values onto the lists of topics an algorithm spits out is by no means a bad thing—but it is in stark contrast to the company’s claims that the trending module simply lists “topics that have recently become popular on Facebook.”

There is a lot to unpack here, complicated by a good deal of confusion about what exactly is being alleged:

  • This story is not about the News Feed, that algorithmically-driven stream of content that is at the core of Facebook’s success. Rather, it is about the “Trending News” box of content placed in the upper right of a desktop Facebook page, or more pertinently for most Facebook users, what appears below an activated search box on mobile. It is valuable real estate in the way that all Facebook real estate is valuable, but it is of considerably less importance than what appears in the aforementioned feed. Indeed, I suspect I’m not alone in that before this controversy happened I didn’t even know it existed on mobile at all.
  • Thanks to Gizmodo’s reporting a week ago, we already knew that Facebook has a content team that chooses which trends deserve to be promoted, writes headlines for them, and also blacklists topics (most commonly because “it didn’t have at least three traditional news sources covering it”). Gizmodo added that “Those we interviewed said they didn’t see any signs that blacklisting was being abused or used inappropriately”, and suggested that the content team was being phased out as Facebook’s algorithms improved.4
  • Apparently in response to last week’s story, a former “curator” from the content team and self-identified conservative alleged that conservative topics were sometimes blacklisted; other curators disputed that claim, but all those interviewed with Gizmodo agreed that curators also had the power to “inject” stories into the trending list even if they were not, in fact, trending. Most examples were about Facebook trying to keep up with Twitter in current news, although longer-running topics like Black Lives Matter were allegedly injected as well.

I parse these details for a few reasons: first, it seems self-evident that a team of curators would, in fact, curate; Facebook’s mistake was in its willingness to let people believe “Trending News” was purely algorithmic. Second, there is very strong evidence that “Trending News” has a human component that, like the “balanced” news organizations of old, is by definition subject to bias. Third, the allegation that said bias is actively trying to suppress conservative news is the opinion of one person only (contra Gizmodo’s headline). And when you consider the make-up of the content team — “young journalists, primarily educated at Ivy League or private East Coast universities”, according to Gizmodo — it seems very possible that the second and third points are, per my observation about the conservative critique of media,5 the exact same thing.

The Rise of Alternative Media

As you might expect, the conservative media was all over these allegations; what is most striking, though, at least in the context of the founding of my old paper The Badger Herald, is that these outlets exist at all. The Internet removed the need for things like desks, typewriters, and especially printing presses, making it viable for an entire new universe of publications. And, unlike the news organizations of old who started with a geographic monopoly and worked backwards, Internet-era publications have no distribution advantage (or more pertinently, disadvantage) versus anyone else; the only way to win is to attract more users on the basis of your content.

To that end Internet publications, particularly political ones, have tended to have a very distinct point of view, whether it be Talking Points Memo on the left or Red State on the right — and those are just two examples of many, covering every part of the ideological spectrum. And why not? The truth is that all of us like to read what we already agree with, particularly when it comes to fraught issues like politics, and we’re more likely to return to a site that makes us feel good about our beliefs.

Facebook has magnified all of these trends: not only is content content, regardless of source, but it also tries to give us more of what we (literally) like, or click on, or comment on (in this case I am talking about the News Feed, not the Trending News section). If you like publications and stories that are more liberal in nature, you’ll get more liberal stories and publications in your feed; it’s the same thing with conservative stories and publications, or sports, or music, or whatever topics “drives engagement”, to use the parlance.

The result is that if you are a conservative, say, you are living in a cornucopia of conservative thought unimaginable to those students launching a new college newspaper against the odds in 1969. There are no obstacles to publishing, and Facebook actually tries its darnedest to bring you more of what you like in the name of engagement.

Polarization and Virtual Villages

Late last month Ezra Klein, who has covered the topic of polarization in American politics extensively, wrote in an overview of a 10,000 adult survey done by Pew about politics:

It’s tempting to imagine that rising political polarization is just a temporary blip and America will soon return to a calmer, friendlier political system. Don’t bet on it. Political polarization maps onto more than just politics. It’s changing where people live, what they watch, and who they see — and, in all cases, it’s changing those things in ways that lead to more political polarization, particularly among the people who are already most politically polarized…

It’s easy to see how this could work to strengthen polarization over time. As Cass Sunstein and others have shown, people become more extreme when they’re around others who share their beliefs. If liberals and conservatives end up moving to different places and surrounding themselves with others like them they’re likely to pull yet further apart. And even for those who can’t move, the internet makes it easy to settle in a virtual neighborhood with people who agree with you. Polarization is going to get a lot worse before it starts getting better.

When Klein refers to “a virtual neighborhood” he means Facebook: that is where people live, where they go in the empty spaces of their lives. It is by far the biggest traffic driver to nearly every site on the Internet, and the most-used app of every age group. And it is a company whose executives talked about engagement double-digit times on the last earnings call. It is the metric that matters, the one everything at the company is built around.

This, then, is the deep irony of this controversy: Facebook is receiving a huge amount of criticism for allegedly biasing the news via the empowerment of a team of human curators to make editorial decisions, as opposed to relying on what was previously thought to be an algorithm; it is an algorithm, though — the algorithm that powers the News Feed, with the goal of driving engagement — that is arguably doing more damage to our politics than the most biased human editor ever could.6 The fact of the matter is that, on the part of Facebook people actually see — the News Feed, not Trending News — conservatives see conservative stories, and liberals see liberal ones; the middle of the road is as hard to find as a viable business model for journalism (these things are not disconnected).

Indeed, one could make the argument that an authoritative news module from Facebook would actually be a civil benefit: at least we would all be starting from a common set of facts. What is far more damaging — and far more engaging, and thus lucrative for Facebook — is all of us in our own virtual neighborhoods of our own making, liking opinions that tell us we’re right instead of engaging with viewpoints that make us question our assumptions.

  1. I don’t usually talk about this much, in part because I’ve almost completely changed my politics since then ↩

  2. It’s almost unfathomable now, but print advertising was so lucrative that The Badger Herald, where I worked, actually paid a staff of 100 or so people across editorial and ad sales who put out a free 16~20 page broadsheet five days a week. As I recall, at that time The Badger Herald’s daily circulation was 16,000, and The Daily Cardinal was 10,000. Needless to say both have dramatically cut back. ↩

  3. Per the previous footnote, having been raised in this environment, I know from experience that the idea of a “liberal bias” to the news, whether true or not, has been unquestioned by conservatives for decades ↩

  4. Indeed, as the Huffington Post reported, today most people don’t even see the same topics. ↩

  5. And without weighing in as to whether or not it is justified ↩

  6. And, of course, algorithms, having been created by humans, have their own biases ↩