Venture investor Marc Andreessen famously wrote “software is now eating the world”— encoded intelligence dominates business. A new book (Platform Revolution) wants to update the slogan. Author Sangeet Choudary (and co-authors, Geoffrey Parker and Marshall Van Alstyne) think the new omnivore is really “the platform:” the digitized, open and participative business models creating commercially connected ecosystems of producers and consumers. The networks and markets forming around—and orchestrated by– Google, Airbnb, Uber, and other virtual exchange enterprises are the maws into which traditional companies are now disappearing.
Understanding The Juggernaut
The tantalizing leap of this book is its focus on a broader pattern of technology-enabled value. The revolution the authors sketch goes beyond algorithms (Google), “on demand” resources (Uber), or “sharing economy” databases (Airbnb). It’s all of the above and more. Clever software is the core of every platform business, but the book argues that the real juggernaut isn’t the code; it’s the explosively scalable network processes that are disrupting traditional buy-make-sell competition.
Platforms beat old-fashioned “pipe businesses” (as the book calls them) by building new growth markets and innovating communities beyond the boundaries of their central hubs. While taxi companies and Marriott struggle to market assets they own, Uber and Airbnb are expanding into unbounded universes, joining ever more producers and customers. Google’s platform links an infinite supply of people looking for answers to a near infinite supply of advertisers eager to sell into those needs. Platforms win, not just by facilitating such new interactions, but also by aggregating and analyzing the data of it all — so everybody out there gets better at demanding and providing services to one another. It’s a tsunami of monetized organizational learning and adaptation, facilitated across networks of networks.
Probing The Nature Of The Revolution
Platform Revolution does plenty pitchfork-waving, but also offers sober advice for how to get with the transformation. It’s a well-illustrated learning tour of the economics and social dynamics of winning (and occasionally failed) platform businesses.
I spoke recently with Sangeet Choudary about the book, and the wider research that’s earned him a place as a Top50 Thinker. Six insights from our conversation:
Looking Back
1. We’ve seen platforms before—but now, “this time is different.” “Platform” is not a brand-new idea. Consider, for example, the 1957 “corporate strategy vision” of Walt Disney, diagramming how his creative studio would produce characters and stories to fuel (and be fueled by) a web of businesses in television, music, theme parks, publishing and merchandise.
The authors acknowledge such antecedents but argue we’re now in a whole new game.
Sangeet explained: “Global network infrastructure now allows for much larger scale aggregation, no longer dependent on physical exchange. Markets and other forms of virtual value creation are more dynamic and open than before, and increasingly self-governing—not controlled by contracts or central enforcement. Value itself is more democratized—consumers can be producers and vice versa: you can become an Airbnb provider, but also use the service for yourself when you travel. Finally, the explosion of data, and its use by platform businesses to keep learning is perhaps most significant. Uber, for example, is not just matching rides to travelers, but increasingly predicting and even structuring demand by algorithms that rebalance supply of available cars. Business models are bigger, more virtual, more dynamic, and more intelligent than ‘platforms’ of the past.”
2. Platform value derives from creating scalable, complementary kinds of connections: markets, ecosystems, and communities. Much platform discussion treats markets and ecosystems as interchangeable ideas; and inevitably “community” is another invited guest at the conceptual table. I came away from Platform Revolution wondering if we don’t need more semantic precision. I posed the question to Sangeet.
“Good question and yes, our book could have been a bit clearer about the relationships among these terms. Platforms create an exchange of value enabled by technology, and the market is comprised of participants making the exchange. The ecosystem is something larger. The overall value of a platform usually requires other players too, e.g., developers who build tools to operationalize the exchange. For Twitter, the platform market is comprised of tweet creators, tweet readers, and advertisers. Its ecosystem would also include developers working on, say, the search function for the platform.”
But What About “Community”?
“Value exchange based on markets depends on incentives, and we think about it in economic terms. But there are two additional elements to the model. Platforms also have codes and cultures; these ultimately shape what becomes a community. Codes are rules of conduct that mediators enforce for operating in the market. Culture is more emergent. It grows out of values and practice as the market and ecosystem mature. It’s more difficult to control; and also (as seen with Reddit and Twitter), the culture of the community can sometimes turn against the platform itself.”
So if culture and community are important, but difficult to control, what’s a good platformer supposed to do?
“We’ve seen that different platforms have different degrees of dependence on community. When market exchange is more commoditized—like Uber or Lyft providing taxis to customers—strategy calls for managing economic incentives. But when a business is more individualized and varied, say apartments in Airbnb or craft sales in Etsy, strategy demands more attention to culture across the markets and ecosystem.”
Looking Ahead
3. Platform strategy is evolving towards a second phase of issues, as the concept matures. Sangeet noted how many traditional companies trying to migrate to platform growth often fail: “The first phase is making markets and exchange work. Companies stumble either by pursuing only incremental changes to their traditional business, or by not fully embracing a big enough ecosystem to create new value.”
OK, I said, but what about longer term?
“Brand will become more important, because ultimately differentiation will move beyond creating efficient market connections, towards overall end-to-end experience. For example, Airbnb has to compete long term on that, and not just matching people with nice apartments. That leads back to questions of culture. Over time, even for some businesses that are initially commoditized, community will start to matter more.”
I asked if, given the inevitable competition as platforms like Uber and Lyft go head to head, “winning the game” will become just a new version of Michael Porter’s classic framework of forces and competitive advantage.
“Yes and no,” Sangeet offered. “Some platform ecosystems will competitively differentiate themselves, e.g. Vimeo co-exists with YouTube by providing a different experience and tool set. More directly competitive ecosystems will try to close off their rivals—but the fight will turn on controlling demand (actors making the value exchange) whereas traditionally, in the Porter model, it was about controlling supply. For example, in earlier times, your advantage as an oil company was controlling the sources of crude. Today, platform businesses have to try to gain an edge with people offering and buying services. Maintaining that advantage within an ecosystem is much more difficult. An ecosystem can turn against you—as happened with Samsung and Google/Android. That doesn’t happen with an oil field you own.”
The New Oil
4. Watch for competitive battles in the future about data ownership and value. Speaking of bubbling crude, the book reminds us that “data is the new oil.” Sangeet emphasized some further implications about this dimension of the platform revolution.
“Data will become more of a regulatory issue—because members of ecosystems are essentially subsidizing the collection of—and learning from—data as they participate in platform work. Uber drivers, for example, provide huge amounts of data as they drive around—looking for customers, paying for their own gas as they go. We’ll see more pressure for sharing the benefits of aggregated data.”
“There will also be continuing global political issues about data. Some platform companies may end up knowing more about citizens of a country than the country itself. That’s one of the reasons China is so adamant about maintaining firewalls around its businesses.”
Whither Leadership?
5. Platforms require a different kind of leadership—attuned to the cultures and governance of the ecosystems they are shaping. My only real critique of this otherwise compelling book was its sometimes bloodless analysis—not enough discussion of people and leaders. Sangeet acknowledged the gap—“remember, Brook, it was two economists and a computer scientist collaborating here”—but he rose to my challenge with a few helpful observations.
“Leaders in platform businesses have to learn not to think of people in their ecosystems like numbers in a traditional market analysis. For many leaders that’s hard to do. Future strategy will call for adapting to ecosystems what best firms now do internally—focus on emerging high performers for extra development, and manage others with more cultural incentives.”
“There’s also a leadership skill we see emerging in platform companies, about deciding ‘what’s inside, what’s outside’—judging what work should get done at the core of the business, and what belongs out in the ecosystem. It’s not about cost, but more the best way to create value and experience overall: who contributes what to that, where do they sit?”
“New questions will also arise as more tools and infrastructure allow ecosystems to become truly self-governing. Firms, as we know them today are, at least medium term, not going to disappear—but they are starting to shrink in size, as platform models become more important. The next phase of this revolution will raise a whole new set of leadership issues.”
Originally published on Forbes.com