The Complexity of Platforms

Platforms are organizations or infrastructures that enable networks. Over the last decade social and digital platforms have emerged that have replaced or amplified a variety of institutions, markets, and organizations. As platforms successfully scale up the number of their participants, they are able to reduce the cost the participation, increase the value of participation, or both. The rapid value creation associated with platforms often shift or create new the centers of power in the systems they moderate.

Due in larger part to disjointed improvements in telecommunications, platforms are increasingly represented among the organizations that impact our lives. Well known commerce-oriented platforms include the digital distribution of consumer goods (Amazon Marketplace), hotels (AirB&B), travel (Orbitz), music (SoundCloud), collectibles (eBay) and transportation (Uber). Such platforms have transformed, and in many cases completely replaced, long standing markets. Less well publicized are institutional platforms, such the World Economic Form, the MIT Media Lab, and the Santa Fe Institute. Like commerce-oriented platforms, the utilization of networks enables institutional platforms to impact the world on a scale that is surprising given their small organizational size. To date, the most impactful platforms may been in the space of information dissemination, where examples include Google, Wikipedia, Slack, Reddit, Stack Overflow, GitHub, Twitter, Facebook, Instagram, Tumbler, and Snapchat.

In every case successful social and digital platforms include some number of key ingredients matched with clear, and often immediate, economies of scale. These include:

1. An essential or core utility
2. Mechanisms of distributed coordination
3. Mechanisms of diffusion
4. Powers of computation
5. Open-ended scaleability

Taking these in terms in order.

An essential or core utility

Platforms begin with a well defined function, and seldom include in their initial release or incarnation, all of the institutions and market that they ultimately influence.

Wikipedia was first and foremost an online encyclopedia. Uber was first and foremost a taxi service, and Amazon an alternative to the physical bookstore.

Mechanisms of distributed coordination

Platforms all exploit the internet and the web to move the burden of coordination to individuals and away from an unscalable hierarchy. Essential features of platforms make local contributions and consensus formation possible. An analogy was the establishment of universal currencies that overcome of the coincidence of needs served by barter.

Mechanisms of diffusion

Platforms make any one source of information available to all users of the platform. In this way a platform behaves much like a circulating currency or information or trust. When coupled with desirable properties of networks, such as the small world property, then all individuals can reach all others with very few intermediaries and no need for a central server.

Powers of computation

Many platforms are founded on computational models and thereby allow computational simulations to amplify their core functions. This is why Netflix and Amazon can add a recommendation systems to what is otherwise merely a ubiquitous store front, Uber and Google Maps can provide analytics on traffic, route economics, and providers in addition to their core utility, and Google Ventures can mobilize the power of data analytics to guide investment decisions.

Open-ended scaleability

Despite the value of their core utility, coordination and diffusion functions, and computational services are, most platforms would not survive without a model that is scaleable. The lack of scaleability is frequently the cause of platform failure. From the debacle of Maxis online game Spore, and the self-imposed constraints of the search engine Go.com.