The Science of Digital Platforms
Digital platforms are evolving. They used to underpin business functions, then they ran whole businesses; now, the most advanced digital platforms facilitate business ecosystems, as we see in hyperscalers such as Facebook, Apple and Microsoft. If a traditional organization is prepared to go through significant change, an advanced digital platform can give it the ability to constantly evolve – and even to beat Amazon.
This report builds on the ‘routes’ to the platform model introduced in our 2017 report The Renaissance of the IT Organization, outlining how these new digital platform environments change the operating model of the whole organization, to one that can evolve fast and enables constant evolution. As we described in our 2018 and 2019 reports Liberating Platform Organizations and The Platform Explorers’ Guide: Four Paths to Platform Business Models, this is the advanced operating model of today’s leading software companies and platform businesses. In this report we use models drawn from the natural and social sciences to help describe four key approaches used by advanced digital organizations:
- Modern ecosystems
- Data fluidity
- Digital platforms
- Digital platform teams
Inside-out evolved to outside-in; outside-in is evolving to across ecosystem
In the early 2000s, LEF’s David Moschella used the term outside-in to describe how IT used to be an entirely internal matter, but the sensing, strategizing, planning, designing, building and sourcing of IT had all started to be driven from outside rather than inside the organization. Now, in the 2020s, IT strategy and execution are moving again, into modern ecosystems: dynamic constellations of organizations in which we may not be at the centre or even anywhere near the centre. Being part of an ecosystem changes our perspective. Modern ecosystems are more powerful than traditional firms because they can construct customer- or citizen-centric combinations that were not feasible before. But the coordination and orchestration required to do that relies heavily on the flow of high-quality, trusted data – a characteristic we call data fluidity.
Figure 1 – Data fluidity enables dynamic ecosystems to serve the customer journey
Furthermore, data fluidity enables organizations connected to each other in ecosystems to work at scale and learn at scale. At present, these are advantages that only the hyperscalers have. These digital leaders harness data fluidity to drive their data flywheels to generate better insights, which drive improved services, which capture more and better customers, who supply more relevant data to the algorithms. It’s a self-reinforcing flywheel of feedback loops. A few of the hyperscalers go further, improving data fluidity in their ecosystems by sharing critical data with select partners whose own flywheels then amplify the learning and experience within this inner circle. They have industrialized learning and can now compete on experience curves.
Data fluidity enables organizations connected to each other in ecosystems to work at scale and learn at scale.
The most efficient vehicle to increase the data fluidity of the ecosystem is the modern digital platform – with its APIs, analytics, artificial intelligence, machine learning, cheaper processing, edge computing, self-service and (critically) a platform team with a matrix mindset (see below).
The most efficient vehicle to increase the data fluidity of the ecosystem is the modern digital platform.
Figure 2 – The natural & social sciences of data
The natural and social sciences have models that can convey how data works, and specifically the properties of data and the interactions between data, systems and humans. In the report we use simple models from mechanics, genetics, kinetics, economics and ethics to create six strategies to increase data fluidity, both inside the organization and outside in the ecosystem. The foremost challenge is how to solve the veracity of data as it flows around an ecosystem. If there is no trust, there is no ecosystem.
In the report we use simple models from mechanics, genetics, kinetics, economics and ethics to create six strategies to increase data fluidity, both inside the organization and outside in the ecosystem.
Modern digital platforms have evolved beyond recognition
Decades ago, a platform used to be a physical structure – software and data in a computer. Later, platforms became architectures of software and data that ran the processes and controls of an organization (e.g. ERPs), integrating business functions and improving data flows (and data fluidity). Now, platforms have become environments of technology, processes and people that create change through digitization. Platforms used to be things we built; now they are environments for change.
Now, platforms have become environments of technology, processes and people that create change through digitization.
Figure 3 - How digital platforms evolved into environments
At LEF, we are seeing traditional organizations changing significant parts of themselves to operate like software companies – small teams utilizing agile approaches and an advanced digital platform to create or reengineer business processes. The most advanced platform environments (such as Platform DXC) are plugged into high-performing ecosystems and have a mature production platform. This speeds up the transformation of IT-intensive parts of the organization’s operating model. Our observation is that platform-based environments are initially one of three types: customer engagement environments that focus on customer propositions and experience; production environments (not to be confused with traditional technical production environments) that create and renew core technology; and operations environments that focus on processing transactions and managing assets. As these platform environments go through application modernization, application transformation and application innovation, they become more like each other, and progressively change the operating model of the people and processes using those applications to one that can evolve faster, all the time. Their traditional operating model changes to the one used by software companies and hyperscalers.
The platform team is a change agent
As the nature of a digital platform evolves, so does the purpose of the platform team. The mature platform team is a change agent on the organization’s journey to becoming like a software company. To drive that change, the platform team continually improves the performance of development teams, industrializes components and drives component innovation, re-sources components and creates new delivery services.
Figure 4 – The capabilities of a platform team
In order to maximize adoption of the new platform environment and its change practices, the platform team must design, build, sell, manage and operate the platform environment as though it were a product. Typically, it uses Product Management (PM) disciplines matured in the consumer product and software industries. In 2018, LEF published The Matrix Mindset as a model for leaders in business to ensure they are well-positioned to drive change. It identified the eight areas of the matrix mindset shown below. From our observations of mature platform teams, these same eight characteristics seem to be what enables them to build successful modern ecosystems.
Figure 5 - The eight areas of a matrix mindset
The appliance of science: How to start evolving towards a modern digital platform-based organization
The report sets out initial steps for traditional businesses to take:
- Be scientifically systematic:
- Identify the intended and unintended ecosystems your organization is already in
- In any ecosystem, gauge and tune your data interactions
- Form a view on your own and your ecosystems’ data fluidity
- Get scientifically methodical about data fluidity
- Run science experiments to help evolve, organize and deploy your digital platforms
- Use rocket scientists – build out your platform teams