As you’ve probably seen by now, almost any industry can be digitally optimized. Here are a couple of examples.
Salman Khan of the Khan academy wants to do optimize education. He wants children to learn on a digital platform in their own time and at their own speed. In his inverted view of education, the classroom is where children come to review, discuss, and get special attention from teachers on areas where they need help. Because the teacher can see on the platform what the student did and what they struggled with. The students’ performance on the platform is already quantified.
Pathadisha (which means directions for travel) is a very interesting startup I met, which connects all modes of public transport in Kolkata, India, and also tracks users’ data via their smartphones. Consumers can track arrival times at bus stops for any of 18,000 individual buses by route number for example. But collectively, by pulling all of this data together, Pathadisha gives you a snapshot of ferries, trams, buses, subway, and trains, all plying in the city. This is being used by the local authorities to schedule transport better and by private providers to assess when surges in demand might occur. Pathadisha is optimizing public transport and was even used to manage temporary travel passes on the Kolkata Metro (subway) during COVID-19 restrictions.1
The Big Trends
You will have picked up a few common patterns through the book. Let’s look at some of these big trends, which are germane to every industry. Interfaces are continuously evolving—from mobiles, to sensors, to nanotechnology inside our bodies, to chatbots on websites, to voice and natural language, and even to brain signals. Service design frameworks are critical to getting users to adopt these interfaces.
The stream of data from this universal connectivity explodes into a tsunami of data. The continuously falling cost of storage, better tools, and frameworks for dealing with data and improved analytical frameworks and AI both address and encourage this explosive growth of data. For businesses, the worlds of operational data, customer data, environmental data, and enterprise data merge into a single ocean of usable information. The successful future organization will be a data-centric organization, but we are still to understand what that means in terms of structures, practices, and processes. Certainly, data protection and ethics will take center stage.
Work gets smarter—we are freed from the shackles of office desks and working hours, and new ways of measuring work will evolve. Processes will get atomized. Company boundaries will blur, as well industry definitions. This evolution of work will have huge societal implications. We will need to address the many challenges if we want to reap the benefits of a postindustrial phase.
Businesses will need to be architected for change. Not like transformer toys, which can be one of just two things, but like Lego sets, which can be rearranged repeatedly, quickly, and effectively. Cloud enablement, and moving to the Enterprise as a service, coupled with strong API layer management and smart security will be the levers of this change architecture. A lot of organizations will have to frequently restructure and reorganize, but the winners will be those who can control this better and do so quickly. Those that take months for new structures to start making sense will suffer.
Single, monopolistic markets will abound, especially in the digital component of any industry. The scale-free network pattern will drive ecommerce, media, and other industries to this highly skewed model, of winners take all. Regulation will be one of the primary barriers to this monopolistic trends—whether it is regional such as in China, or global as in the case of banks.
Disintermediation and platformization will continue to create and disrupt industries. Efficiencies will be found and removed via platforms and removal of nonvalue-added intermediaries. Providers and creators will be connected more directly to consumers. The successful dis-intermediaries will be platforms such as TripAdvisor or Spotify, rather than resellers or aggregators such as traditional travel agents or record labels.
We will see two kinds of automation—task and process automation, and decision and outcome automation. Task and process automation will typically be done by software and hardware systems that are often described as robotic. These will work within specific boundaries and have a relatively narrow set of functions and conditions under which they work. Cars are being built in robotic factories and a lot of manual processing is being replaced by robotic process automation (RPA) software. On the other hand, decision automation tools powered by AI will start to replace much more advanced work—such as driving cars and managerial roles, even creative ones. AI is clearly a mega trend and could take us into the realm of singularity.
All of this will happen in increasingly smarter and networked environments. The energy network, transport network, and our homes and offices will all be able to share data, and smart devices and processes will feed off that data. Your electric car left overnight to charge will negotiate with the smart grid to agree the best rate and time to charge, thereby giving you the best deal and also allowing the grid to load balance. Our decision models will be significantly more capable of understanding and exploiting network effects rather than just individual behaviors. For example, methods for fighting crime or fraud, and addressing mental health and other health care problems could be significantly improved by the increased modeling of network effects.
All of these changes will place a huge responsibility on the leadership of organizations at all levels. Leaders will need to combine the ability to manage design, technology, and business, appreciate regulatory impact, and continuously craft and update their vision of the world today and tomorrow, and be able recalibrate this journey, and their organizations. The drive to becoming a data-centric organization should become one of the most critical tasks of leaders. Correspondingly, there needs to be a significant re-education process in the board room around data, risk, security, and the accelerating change.
The Conceptual Digital Framework
I’ve always found it useful to structure digital thinking into the layers—technology (interfaces and digital infrastructure), design, data, and analytics. This protects us from confusing and overlapping concepts. Sometimes even one system such as an app on a phone can be broken up into these three layers—especially if this represents a service such as a bank or streaming music. Each of these layers comprises many different technologies, which are all evolving at varying paces. Design thinking is the critical differentiator between the adoption or rejection of new technologies.
Automation, robotics, or AI systems are by themselves categories of systems that have all these layers. We also need to consider that the next generation of digital technologies will merge the digital and physical worlds as never before, so those of us who have grown up distinguishing between the real world and the digital world will have to reorient ourselves all over again.
The Future Is Not More of the Past
Despite the vast and sweeping changes that we’ve seen so far, in the past three decades, it’s sobering to think that we may in future think of this period as one of glacially slow progress. I’ve pointed to two specific points—that is, the invention of HTML (which led to the birth of the World Wide Web) and the creation of the smartphone as key points of inflexion. In future, these points may happen more frequently, or there may be a much more diffused and continuous set of inventions and discoveries that drive us forward faster than ever. Exponential change is all around us.
Transformation is clearly a huge underlying theme here. The role of technologists may go from building and operating technology, to driving organizational architecture and governance, and having the data stewardship in connection with the data and security officers. In all this, the need to handle exponential change while retaining resilience will still rest with the leadership and the board.
How Do We Make It Work?
Our simple methodology, therefore, which calls itself connect, quantify, optimize is a really an adaptation of the basic scientific method. Whereas, in a scientific experiment, you might need to gather data explicitly, digital by its nature allows us to gather the data out of the very interaction of the user with the technology. The analysis of this data can lead to a restatement of the whole business premise—known in the startup world as a pivot, or it can be a marginal tweak to the solution to make it a little bit more effective. This cycle of connect, quantify, optimize, or to restate it: try, learn, evolve can be done very quickly and efficiently with the right tools and skills in the digital space. But it can grow rapidly to redefine large global corporations as well.
Let me end with a story. One morning about three years ago, we came downstairs to discover that our car had been stolen. Our neighbor’s CCTV footage showed a couple of men getting into the car and driving it away at 4 AM. As it happens, just the same week, we had signed up for a new insurance for the car, which involved putting a tracking device on the car and using the data to rate our driving and give us a better quote. We promptly called the insurance company to report the stolen car but also to tell them to track it using the device they had put in. The first response was that the telematics team has a 14-day response SLA. Then we were told it was outsourced to a different company. Later in the day, thanks to my wife’s persistence, it emerged that it wasn’t a separate company but a different division. From Friday morning, till Monday, we pestered the insurer until at 9:30 AM on Monday we got through to the telematics team. Who took 30 seconds to tell us exactly where our car was. We spent all of Monday staking out our own car until the police arrived to take over, and our car was finally recovered safe and sound.
What’s really interesting is that neither the insurance provider nor the police thought of the tracker as the obvious and immediate option. We were actually covered for the cost of the car, so the loss would have been the insurers. No matter how great your technology is, unless you reshape your operations around it, you won’t get anywhere near the full benefit from it.
That’s really the bottom line to this book. Technology evolution is fundamental. Design thinking is crucial, and data and analysis are essential for digital change and success. But none of them by themselves is sufficient. Combining them and redefining the way you work, or even what business you run, is the only way you’ll succeed, or perhaps even survive in the digital age.