How the Internet Evaporates the Middle: Unbundling Media

#1 in a series of four articles from eParachute.com:

• Unbundling Media
• Unbundling Higher Education
• Unbundling Inequity
• The Future of Work and Learning

I’m frequently asked by friends who aren’t in hi-tech: What just happened? Traditional media companies are gone. Industry after industry is being changed dramatically by technology: Is no business safe? College is unbelievably expensive: Will it change as well? What are “incubators,” and why are there so many of them? Will Google and Facebook run the world? Why have so many older workers been unemployed for so long? What’s happening to the middle class in America?

And… are all these things somehow connected?

——————————————————

This is a story about The Middle.

Or, rather, about the disintegration of The Middle. Across industry after industry — from music, to magazines, to newspapers, to books — the Internet has provided opportunities to new businesses large and small. But it turns out that, in many (but not all) industries, the Net — and hi-tech in general — has effectively evaporated The Middle.

What do I mean by The Middle? Before the advent of the Internet, mature business markets could be regarded as pyramids, with organizations occupying every available segment — from small, local, niche businesses at the base, to dominant, multi-national conglomerates at the top of the food chain — and everything in between.

That “in between” was critical. Not every company could grow in scale from small to mid-sized: Some are born small businesses, and stay small businesses. But even though many markets historically experienced periods of consolidation, most could be relied to support a healthy ecosystem of players — and in the past, it was considered quite possible in many industries to be a mid-sized player.

At least, that’s the way things used to work. Then along came the Net, and businesses like music, newspapers, magazines, and books were all blown quite literally to bits. The result was that the middle area of markets that were formerly occupied by mid-sized players became wastelands.

Your first question might be: Is that really true? I can’t stun you with statistics defining exactly what The Middle is (or was) in any given industry. I’d love to hear from researchers who can provide the data to support or refute this in specific arenas.

However, I’ve found that people in a wide range of industries — from media to manufacturing, from telecommunications equipment to food production — have immediately agreed with the basic premise: The Middle in their industry has evaporated, and the Internet and related technologies are the culprits.

Why is that? What are the dynamics of the Internet that made it the perfect vehicle to disrupt the center of existing markets? If we understand how these traditional markets got blindsided by the Net, we can make some pretty good guesses about the next industries to be disrupted – and how they’ll evaporate in The Middle. And along the way, we’ll get some insights into what’s happening to global economies, including the disruptive changes in the way we learn and the way we work.

To start, I’d like to weave together several themes that have become part of the public discourse: The “unbundling” of businesses, the true dynamics of the so-called Long Tail, and the already-disrupted dynamics of the media industries.

Long Tail, Evaporated Middle

I received my first lesson in the impact of the Internet when I was editor-in-chief of Network Computing Magazine in the early 1990’s. I encouraged the editors to write a cover story on “The Internet Gold Rush” in 1994, soon after the release of Mosaic, the first real Web browser. We got a number of things right, including the disruptive nature of this new communications infrastructure, and the inevitable use of the Net for collaboration. But we also were way off the mark on a number of predictions, such as assuming that online advertising would never fly because of the prevailing pioneering ethos that defied commercialization. (In fact, we called ourselves “netizens” back then — Internet citizens — a term that’s now long-buried in the fossil record.)

I was wrong with another prediction. While it was clear that the Internet would disintermediate many existing players, I believed that the rising tide of technology would essentially lift all future boats, allowing healthy business ecosystems to thrive. As former Wired editor Chris Anderson popularized, like any power law graph, there would be small and successful companies in “the long tail,” and large and successful organizations in “the short tail.” So doesn’t it make sense that all those in between — which we could call the mid-tail — would be successful as well?

It turns out that in many industries there is indeed a long tail, and a very healthy short tail. But there isn’t much of a mid-tail any more.

The classic framing for this is that these businesses used to produce things you could touch (atoms), and now that their output is completely digital (bits), they behave in fundamentally different ways. What’s clear now is that the reason Internet-fueled industries behave so differently from before has to do with what has classically been called “unbundling.”

Unbundling Unbundled

One of the first publications I’ve seen on the idea of unbundling was in a 1999 Harvard Business Review article by John Hagel III and Marc Singer, which explored the unbundling of the corporation. But the earliest and most classic case of unbundling came in the form of the original clone PCs.

Consider the IBM mainframe of the 1980’s. You may remember these: Massive computing systems so big, they needed huge, air-cooled rooms to hold them. “Big Blue” (as IBM was known) blocked out the light for competitors in the mainframe market because of a relentless strategy of vertical integration. The computer’s hardware (the box), microprocessor (the brains), operating system (the management software), and applications (what users ran) were all tightly integrated together. If you changed, say, the microprocessor, you needed a new version of the operating system and, often, new applications to go with it. Choice? Sure. You can have any color you want, so long as it’s Blue.

This approach to market domination is known as vertical integration. IBM was by no means the first company to block out the light for competitors with this strategy: Carnegie Steel did it in the late 19th century, by owning businesses that covered every step in steel sourcing and production, from digging up raw materials to delivering finished girders. And petroleum companies such as Gulf Oil did it in the 20th century, from oil exploration to delivering car-ready gasoline at the corner station.

In technology, vertical integration gave great market power to those with scale and brand like IBM, making it very difficult for new entrants to come into the market. Of course, other companies could run their application software on IBM’s hardware — if they were willing to pay hefty licensing fees. New competitors arose, offering minicomputers and workstations that chipped away at the mainframe market. But these were vertically-integrated systems as well.

Then along came Intel and Microsoft, and computers were suddenly unbundled.

Intel allowed software developers to create programs that could send instructions directly its microprocessors. Using those instructions, Microsoft’s operating systems such as DOS and Windows in turn supported the software applications from other developers. That allowed programs like Visicalc and WordPerfect (the Excel and Word of their day) to run on any computer using Intel’s processors. If Intel changed the processor, so long as it continued to support many of the existing instructions, Microsoft’s operating systems and the applications that supported them would run just fine. And though hardware manufacturers like Dell would integrate all these pieces, and sell them as a bundle, you could buy a completely different set of components from another manufacturer, and the same software would still work. (At least, that was the theory.)

Because all of these components — hardware, the microprocessor, the operating system, and the applications — could work independently, they became unbundled.

Computer science types call this kind of role definition for various technology components abstraction layers: sets of functions typically performed by technologies in different categories. Things that run at different layers work together because there are software instructions called APIs, or Application Program Interfaces, that allow various kinds of software and hardware to work together interchangeably.

By allowing these components to work interdependently, hi-tech companies were free to innovate independently. Competitors at each “layer” arose, created “clone” microprocessors, competitive operating systems, and new applications. Companies could innovate in parallel, moving the technology forward rapidly at each layer.

Unbundling doesn’t necessarily last forever: Successful technologies often consolidate into a small number of hands, and dominant companies can exert significant market power. Microsoft ironically “re-bundled” the software part of the PC back together, selling productivity applications like Word and Excel in a “suite.” And by already controlling the operating system market with Windows, Microsoft exerted control over the market even more completely than IBM had.

Until, of course, the Internet came along.

It turns out that the Internet is the ultimate unbundling machine. By allowing applications to be run in a Web browser, applications could run relatively independently of the underlying operating system — or, in the case of mobile applications, without a browser at all. When Microsoft co-founder Bill Gates famously wrote “The Internet Tidal Wave” memorandum in 1995, forcing Microsoft to change course to respond to the rapid growth of the Net, it was specifically because he saw the threat that a browser-based world posed to his company.

As it turned out, the Internet is an extremely efficient unbundling machine. And it doesn’t just unbundle software applications and other technologies: It allows for the unbundling of entire industries. And the Internet’s “canary in the coal mine” — the industry that was the first to be disrupted — was the media business.

How the Internet Unbundled Media

The media industry’s core business model can be defined as “content-driven attention.” From an elementary-school newspaper, to AARP’s magazine (the largest-circulation publication in the world), and everything in between, the process is the same. Someone creates content, and if it resonates with readers or viewers, an audience is attracted to that content. If consumers will pay enough to cover the production and distribution costs of that media, they can avoid ads. But if content consumers aren’t willing to pay enough for their media, those audiences are packaged up and marketed to advertisers.

To provide this content, and to deliver audiences to advertisers, traditional media businesses were vertically integrated, just like IBM mainframes. Take, for example, magazines. The vertical layers of the magazine business are pretty clear: Production. Content creation. Content filtering and editing (often called “curation”). Distribution. And, of course, Ad Sales.

You bought a print magazine because something in it resonated with you. Whether you knew the work of the writers or not, some faceless editor behind the scenes would choose what to publish. But you didn’t mind the fact that someone was packaging it up for you: In fact, you liked it because it’s a package, rather than a random collection of individual articles. And whether you knew it or not, because you were attracted to that kind of content, you probably fit some kind of demographic or psychographic that a particular kind of advertiser wanted to reach.

Back before the dawn of the Internet, you could start a small magazine, and – depending on the size of its potential audience – grow it to a mid-sized publication. You’d probably start a few more magazines, and eventually you’d have a mid-sized business.

Of course, there are no guarantees you’d continue to make it in The Middle. The costs of “smearing ink on dead trees,” as it’s been called, then shipping it around the country, often made it challenging to make money. National distribution (which often brings with it things like slotting fees in airports) is costly, and making big commodity purchases like paper can dramatically impact a mid-sized magazine’s profitability: Yet it was not uncommon to grow a business through the traditional phases of Small & Niched headed toward Large & Dominant.

Then the Internet quite literally blew the magazine business to bits, and all of its major elements became unbundled.

Suddenly, startups were able to innovate in each layer of the business. Distribution shifted into the hands of dozens of players, starting with Google’s search engine. Content Creation and Production became the playground of publishing tools such as TypePad, Twitter, and even Facebook — allowing anyone to publish. Editing — the process of defining and curating what gets seen by whom — became crowdsourced through services like Reddit, Pinterest, Tumblr, and Flipboard. Advertising split in two, with text and display ads owned by Google, and want ads taken by Craigslist.

We — you and I — created this unbundled world with our attention. We stopped buying music albums, and instead bought songs one at a time from Apple’s iTunes store, or Google Play. We stopped buying many books at our local bookstores, and instead bought them on Amazon.com. And we largely stopped reading magazines and newspapers, and started getting our news online.

In 1999, the National Directory of Magazines listed 31,570 titles. By 2006, the number had dropped 40% – and the biggest hole was in The Middle. Even large companies with substantial resources began to find it challenging to create a mid-sized publication. For example, in 2007 Conde Nast attempted to launch Portfolio, investing at least $100 million to try to create a mid-sized publication, and folded it within two years.

For media businesses in the Internet era, life is tough in The Middle.

Unbundling Evaporates the Middle

By removing the historical market value of vertical integration in a range of industries, the Internet created a dynamic, constantly-changing landscape of technologies that continues to supplant previously-dominant businesses. But how does unbundling evaporate The Middle?

Bundled media companies maintained a number of natural competitive barriers. Hiring great writers ensured quality control — especially if the brand of the publication was pre-eminent, making the writer less powerful. Printing and shipping your own product meant you could get it into readers’ hands quickly — and more cheaply than competitors who didn’t have your scale. And direct relationships with advertisers meant lock-in for the companies who would pay aid the most money to reach a reader.

But unbundling the media business removed those competitive frictions. Readers could suddenly find high-quality content from a range of sources just by Googling a topic. And the automation of ad delivery made it challenging for publishers to maintain direct relationships with advertisers.

In this near-frictionless business environment, small, niched publications at the low end such as Web sites and newsletters can thrive, because they remain close to their audiences, they can keep their costs extremely low, and they can offer small advertisers access to niche markets. At the high end, large, multinational organizations can survive, and even thrive, by leveraging economies of scale, even as their historical profitability becomes challenged.

But a media business that occupies The Middle of the marketplace has substantial challenges. It’s large enough that it no longer is as close to its audiences as its smaller competitors, but it can’t move as quickly as the little guys to take advantage of new opportunities. It also has a range of costs that are difficult to reduce through economies of scale, such as slotting fees for magazines at airports, and the salaries of people in the content production process. The result is that these businesses “die the death of a thousand cuts,” as the Chinese saying goes: As they lose attention from above from thousands of alternative sources of content, and they lose revenue from ads made available from other sources, midsized media companies often become far less profitable businesses.

Even if some media companies might survive for some time in The Middle, many of these businesses still disappear because they have to huddle together for warmth. They either merged to become big players, or were acquired by big players, so they could achieve the economies of scale denied them in The Middle. And not just in media: The Middle is increasingly becoming a wasteland in a broad range of industries — including hi-tech.

Fat-Bottomed World: Why the Middle is the New Chasm

How can that be true? News articles are filled with a seemingly endless parade of startups. Why aren’t there numerous new entrants biting deeply into the profits of incumbent hi-tech companies?

There are multiple factors at play, but one of the greatest influences is Venture Capital. Or, more accurately, entrepreneurs who choose to take funding from investors who follow a traditional venture model. If you take money from a venture investor, you’re basically agreeing that you’re going to sell your company at some point – either a piece of it if you go public, or all of it if you’re going to be acquired. And since the market for Independent Public Offerings, or IPOs, will never again see the pace of its heyday in the late 90’s, the most likely “exit” — the main way that investors will get their money back — is for the startup to get bought. As a result, the startup founder has substantial pressure to sell out long before crossing The Middle.

The situation for those investors has changed as well. Startup companies used to require expensive computing equipment, and office space to house their expensive employees. Software development went relatively slowly. And, as author Geoffrey Moore pointed out in his iconic book, Crossing the Chasm, proving a new product’s viability with customers took a long time, so a relatively small number of startups were able to attain enough sales to customers to achieve market viability before they ran out of money. As a result, the average startup needed a substantial amount of capital to even hope to “cross the chasm” on the way to becoming a midsized company.

Today, however, startups don’t buy computer servers: They store programs and data in the cloud. Software can be developed in a rapid and agile manner. Startups need relatively fewer employees — who can initially work as a virtual team — so the startup doesn’t need a large office. And depending on what the startup is offering, it can even get initial responses from customers pretty quickly to determine if it has a viable market.

As a result, you don’t usually need a lot of money to get your startup off the ground. That’s why there are so many business incubators like 500 Startups, accelerators like Y Combinator, and co-working facilities like The Impact Hub are popping up: These lean startups need to move quickly, so they require advice and mentoring to get to the point where they can test customer responses as rapidly and cheaply as possible. And they all need to move quickly, because it’s relatively easy to copy someone’s idea and get a new product into the market.

That’s why there are many more startups, each requiring a fraction of the investment capital they once needed. Early-stage investors who used to put larger amounts of money in a few companies now put small amounts of money into many companies. It’s only after a startup has established its market, and needs to grow rapidly, that it takes large investments.

Think of today’s startup as a sea turtle egg. The mother sea turtle crawls up on the beach, deposits her eggs, then staggers back into the waves, exhausted. Thousands of baby turtles, left on their own, hatch and instinctively head toward the water. They’re attacked by birds, land mammals — and, for the lucky few who reach the sea, fish. Only a tiny percentage of the turtles grow to adulthood.

That makes this the pre-Cambrian era of startups: An explosion of innovation, with new species being created every day — but a Darwinian process for weeding them out that ensures only a few will survive.

This dynamic also explains the coming explosion of crowdfunding on platforms like Kickstarter and Indiegogo. For many products, crowdfunding is actually a market test: If enough people will pay for just an idea — since the founders typically haven’t even started to develop a product when they begin their crowdfunding campaign — the company has already proven there’s a viable market before shipping a single SKU. Sure, the nascent business now has to prove they can actually produce, ship, and support the product. But they’ve already built customer centricity into the company’s DNA, which is a pretty good attribute for any startup.

And there will be many more crowdfunding platforms. The JOBS (Jumpstart Our Business Startups) Act has lowered the standards for what were formerly known as accredited investors, reducing friction in the funding arena, and inevitably unleashing a flood of new crowdfunding platforms. The good news is that this will continue to fuel a massive wave of grass-roots innovation. But only a tiny percentage of these crowdfunded startups themselves ever become more than small businesses, much less mid-sized companies. And The Middle will remain a wasteland.

You might argue that this kind of consolidation has happened in every mature industry since the dawn of time. That’s true — but that was when the output of the world was mostly composed of atoms, not bits. Now that much of the value being created is digital, the dynamics of consolidation have completely changed. And one of the key drivers is “paranoia.”

The New Era of Paranoia Evaporates the Middle

In the past, incumbents in various industries faced author Clay Christenson’s innovator’s dilemma, a mentality that encouraged arrogance and decision-making paralysis in large corporations, making them patently unable to respond nimbly to new market threats. In the 1980’s, companies like Digital Equipment Corporation and Wang Laboratories came to dominate the minicomputer market, offering mid-sized computers and workstations that took market share from mainframe companies like IBM. (In fact, Digital and Wang were the original “middle-dwellers” of the computer industry.) But Digital and Wang failed to understand the power of IBM’s PC, seeing it as a toy, and realizing too late that they’d missed the massive disruption in their markets that ultimately destroyed their companies. (In a June 2014 article, New Yorker writer Jill Lepore questioned Christenson’s core thesis of disruption. But the data remains strong in pointing to the difficulty of maintaining market dominance in a constantly-shifting landscape — a pace which the Internet has significantly accelerated.)

Even in the fast-paced computer industry, incumbents allowed new entrants to rapidly grow to mid-sized companies, ultimately becoming super-sized threats. In the late 1980s and early 1990s, a Utah-based company named Novell dominated the computer networking industry, and another Utah-based company, WordPerfect, blotted out the light for other productivity applications. Even though the two companies eventually merged, huddling together against the onslaught of Microsoft, they eventually became part of the hi-tech fossil record. It would seem likely that this would continue to happen over and over again in the tech industry, and new mid-sized players should eventually be eating the incumbents’ lunch. (In an ironic twist, Novell’s former Chief Technology Officer, Eric Schmidt, today is Google’s executive chairman.)

But the leaders of the Googles and the Facebooks of the world clearly understand these lessons. They know the innovator’s dilemma has repeatedly taken down former behemoths. As a result, they have designed their corporate cultures to remain fast-moving and highly aware of what’s going on in their industries. For many of these now-giant companies, former Intel head Andy Grove’s mantra — “only the paranoid survive” — trumps Christenson’s clueless incumbent.

How do the new technology giants do this? First, they continually scan the market to identify startups that could potentially eat their respective lunches. Many large tech companies have created extensive sensor networks — their venture investment and acquisitions arms — designed to uncover and then buy smaller competitors, long before those upstarts grow big enough to become true threats.

The litany of rapidly-acquired companies continues to grow. Look at picture-sharing platform Instagram, purchased by Facebook for about $1 billion. Or Tumblr, bought by Yahoo! for about $1 billion. Or commute-sharing platform Waze, purchased by Google for, yes, about $1 billion. Or Facebook’s acquisition of WhatsApp for a whopping $19 billion.

Could any of these startups ever have become a mid-sized company? We’ll never know. Because they filled a key product hole, or simply represented potential competition to some core offering by an incumbent, these companies were taken off the table by bigger players.

For our sea turtle analogy, the incumbent tech companies are like the fish in the ocean: For the few that can stay alive long enough to reach market viability — crawl into the sea — a variety of voracious finned predators are waiting.

Another dynamic that’s different today is that tech incumbents are increasingly willing to cannibalize their existing businesses by developing or acquiring new products. There’s a clear rationale for this: If a startup could create a better product than you, why not build that product yourself? Sure, there are cultural hurdles to overcome — how does a product manager from a successful product line feel about competing with her fellow employees? — but the company overall wins if the new product line successfully overtakes the old.

Call this “the innovator’s advantage” — Internet style. If the author of The New Technology Elite Vinnie Marchandani, is right — every company today is a technology company — then incumbents in every industry are likely to start thinking and acting in this manner, if they have the management and corporate culture that allows them to do so.

Finally, midsized hi-tech companies that find themselves in The Middle are increasingly turning out to be stagnated companies unable to sustain long-term growth, and therefore need to either merge or become acquired. A McKinsey study states the challenge succinctly: Grow fast or die slow.

Beyond Media: The Next Great Unbundling?

The Internet has effectively unbundled most media businesses, including magazines, newspapers, music and books. The television industry is being unbundled as we speak, and the in-theater film industry is starting to experience the effects of being unbundled. It’s important to point out that unbundled industries don’t disappear completely: Some number of existing businesses survive, and a few might even thrive, albeit as smaller companies. Radio, TV, and movie theaters should all be around for a long time. But overall revenues of incumbent businesses in each of these arenas is likely to be comparatively less than they were before their industries became unbundled.

Given the dynamics of The Lost Middle, there’s already a lot of information pointing us to the next likely market to be disrupted by The Net. It has a number of characteristics similar to media, but it’s much bigger — almost half a trillion dollars. The initial fractures are occurring right before our eyes. And the disruption of this industry will transform our society like no other shift before it.

Next: Unbundling Higher Ed

Gary A. Bolles, eParachute, gbolles@gmail.com

eParachute is a San Francisco-based startup providing online career decision-making and job-hunting tools inspired by the work of What Color Is Your Parachute?, the most popular career book in the world, with 10 million copies in print. Learn more at eparachute.com.