The term “freeconomics” was invented to describe the point at which the spread of nuclear energy would make electricity “too cheap to meter”. We have yet to reach that point in the energy industry, but a similar scenario is not far from reality in other industries. The cost of processing power, the storage, the bandwidth, and other technologies connected to the digital revolution, decreases following the so-called Moore’s law. As a result, the computing power is rapidly becoming “too cheap to meter,” making it possible for some businesses to offer digital services for free and harvest value in other ways. Facebook, for instance, has billions of monthly active users, none of whom pays anything to enjoy its services. Facebook makes money by allowing businesses access to users’ newsfeeds, through advertising and business pages. Google and all its services, from the search engine to Street View, from Translator to Google Scholar, operate in the same way— advertisers pay the cost of making those services available to users for free. These companies leverage an old business model, dating from the beginnings of the modern newspaper industry: advertising. The advertising-based model is a specific type of two-sided platform approach in which free (or almost free) services are delivered to large numbers of consumers, who then become targets for the advertisers who pay to access the eye-balls of those users. In terms of business models, this approach can be classified as an orthogonal platform with a “client-as-a-target” strategy. It is still one of the most popular value-capture systems in the world of digital services and the main revenue-generation mechanism for the majority of the digital giants. As this approach proliferated across the Internet and app field, customers started to expect all digital services to be free—even those they might pay-for in the physical world. For instance, TomTom and Garmin, both manufacturers of popular GPS units sold for hundreds of dollars in the physical market, tried to move their business to the digital world selling navigation apps for smartphones- but were basically shut out of the market by free navigation services like Google Maps or Waze. To escape this equation “digital service equals free services”, new models for value capturing arise such as: freemium models, which mix free basic services with paid premium ones; or in-app purchase models, which give the basic app away for free but offer users opportunities to enrich their experience with it through paid add-ons; and cross-selling models, in which the free app goes along with a physical product to be sold on the market, such as Fitbit’s companion mobile app to its physical fitness tracker. These models offer new ways to capture value from innovations in a world dominated by “free,” but none of them fits comfortably into an environment in which customers are increasingly used not to pay for what they get. Advertising is a valuable alternative for sure, but advertising companies must then sell their services to companies with revenues, otherwise the game doesn’t work. We cannot expect advertiser to pay for everything we use in a free world. Thus, the question of whether there is a sustainable way to offer digital services for free, besides advertising, becomes critical. One potential answer to this question lies in the considerable quantities of DATA that customers generate using those digital services. A number of companies begun to explore the value of this data. For example, in 2009, Twitter started selling expanded access to its database of tweets to third parties, many of whom generated insights from it through sentiment analysis to understand the general mood or reaction to events or politicians from the tweet stream. The Twitter Political Index, presented during the 2012 US presidential election, is one example of the potential for Twitter’s data stream to support powerful insights. In this way, the company created a new and powerful revenue stream. Google, Facebook, and many others have experimented with similar approaches, capturing economic value from the data generated by users. Many questions still remain…How is it possible that digital companies capture such a huge value through data? What are this data? Where are they coming from? January 9, 2007. In this day we can position the beginning of a great revolution, which in a few years has led to a world very different from the one we knew. That day, during the MacWorld conference Steve Jobs presented the iPhone, the Apple's first smartphone. We can consider that day as day 0 of a revolution that is still going on. The iPhone, but more in general all smartphones and the various smart-devices that followed, are the enablers of this great data-driven revolution. Smartphones transformed phones from a tool for making phone calls to a concentration of sensors as camera, GPS, accelerometer, microphone, touch screen and so on, that enabled the generation of different types of data in a single tool. Steve Jobs, sensing the potential of his new product, launched shortly after the App Store. Doing so, he allowed anyone who wanted, to develop new services - apps - that could enrich the ecosystem of services and functionalities offered by this new product. By doing so, in few years, the smartphone has become a real gateway between the real world - which it studies and analyzes through its sensors - and the virtual world, in which it reports the results of these analyses. A few examples? Waze, the App that played a fundamental role in the spread of smartphone navigators, they transformed the concept of navigators. From a tool you use only when you have to reach a place you don't know, to that tool that helps you to reach any place - even those you know very well - in the fastest way and avoiding traffic. This change was possible because of data collected through the sensors on the smartphones. The accelerometer and GPS actively monitor the speed of the car and, together with the user's own signals and the data coming from other users on the same road, allow Waze to generate real-time and personalized suggestions to reach any given location. Another example is fitness tracking services, such as Runkeeper or Runtastic. These apps track users as they run, analyze their pace, speed, breaks and develop customized training plans that meet the individual's needs. All of this, which has become increasingly common in various digital services, is made possible by combining three different information sources: First, end user inputs: like when we tell Waze where we want to go, or when we tell a fitness tracker how many miles we want to run, our weight or our height. Second, the data coming from the smartphone and its sensors: for example, the speed in the two previous examples, or our current geographical position. Third, the data coming from external sources or third parties that can enrich the previous information, for example the temperature or the percentage of humidity while we are running, which are recoverable thanks to the position given by the GPS that is on the smartphone. All these databases that, thanks to algorithms, are based mainly on artificial intelligence, can give life to a sort of limitless personalization, in other words they allow millions of users to have a highly personalized service for free. All we have to do to trigger this magic is to give away a small amout of our data to the service provider. We can identify two main variables in these services, which allow us to have a classification. On the one hand, the degree of interaction and participation, in other words, how active the user must be in order to have a personalized service. On the other hand, the degree of user contact, when the service observes - even in the background - the direct behavior of the user. If both of these variables are low, we have "Background Monitoring Services," such as Nest, the smart thermostat that in the face of our few inputs - the desired temperature - creates a personalized profile of the heating of our home by observing our behavior (when we are home, at what time we go to sleep, and so on) and claim to reduce our expenses of nearly 20%. A second group is that of threshold-call services, such as Credit Karma, which observes the user's financial behavior and only requires the customer intervention in some specific situations. Both variables are high instead in the case of Digital Coaches. The algorithm observes in real time the user's behavior to adapt the user experience at that moment. As in the case of Duolingo that personalizes the experience of learning a new language. The last case is that of Virtual Real Seamless Services, with a low degree of user involvement but a high degree of background observation, as in the case of Waze. Clearly highlighting this continuous cross-reference between the real world and the virtual reality through data. This classification shows us how data already enabled so many different services, involving the user in a more or less direct way but offering highly personalized experiences. As we said at the beginning, these services are often free or almost free, how is this possible? What are the value-capture mechanisms that these companies put in place? As you might imagine, the answer lies in between the world of data and platforms, in particular orthogonal platforms with a client-as-a-source logic. In the next videos we will explore various of these strategies.
