benefits, some people have suggested that we tax machines to compensate those who lose their jobs as a result of the digital transformation. What do you think about that?
Florian: Taxing machines and technology would be the wrong approach. All long-term economic growth ultimately comes from the fact that machines make us more productive and that we can have ever more specialized division of labor. We can afford the things that make our lives better and easier, because technology enables specialists to produce them for us, and the market allows us to trade them in exchange for our own services in the domains that we are good at.
Alexander: Today, many knowledge workers spend time analyzing data, creating slides, and writing emails. In the next 10 years, what do you think our work will look like?
Florian: It will be more data analysis, fewer emails, and about the same amount of time creating slides.
The more data analysis part is probably obvious. We will be inundated in data, whether we like it or not. The different steps of the analysis process will be more and more automated and commoditized so that even people with very limited experience in data analysis will be exposed to it.
The fewer emails part is perhaps wishful thinking, but I am hopeful that other forms of digital communication will replace it. Tools like Slack, Chatter, Quip, and Teams are great if used correctly.
Their greatest value is that they allow us to organize conversations around topics or projects, whereas emails are organized by who sends what to whom (and who is left off the recipient list). The benefit of that cannot be overstated, as it creates true transparency and free knowledge transfers within organizations. This only works, however, if people are intentional about this. I have seen it happen that a social collaboration tool was implemented and then employees used it to directly message their colleagues, just like they would with email.
One thing I can see happening is that we will use data dashboards more often in meetings rather than slide decks with static charts. Charts and figures are often the cues for further questions. My colleague Andy Cotgreave likes to say that the quality-control method of asking “five whys,” as pioneered by Sakichi Toyoda at Toyota, is what one should use when using data to answer questions: don't stop until you have asked “why?” five times.4 Being able to interact with data during meetings allows one to go down such routes of interrogation.
Alexander: Do you think that the automation of processes and the commoditization of digital tools will mean that we will have much less work to do? Would it be an option to reduce work time, let's say to three days per week, to avoid a loss of jobs?
Florian: Individual tasks can take up less time when new digital tools are introduced, but it is a fallacy to think that the overall amount of work will decrease as a result. There is no fixed stock of work that needs to be done. Instead, when new tools help us increase our productivity, we, as a society, tend to produce more. Hence the effect on economic growth I mentioned earlier. Also, don't forget people have to build these tools.
There are two caveats to this answer, though. First, individual jobs will fall away. Companies used to have typists who would type up the letters that their bosses dictated for them on a voice recorder. With the advent of the personal computer, workers can now write their own letters, and as a result many typists lost their jobs. This process will repeat itself continually as new technologies are adopted to automate tasks. Therefore, our education and welfare systems are crucial. They can ensure that people have the right skills for the right jobs and that they can transfer from those jobs that have become obsolete into those that society requires.
Second, we might still see a change in work time in certain professions. The traditional five-day workweek with eight-hour days made sense in the industrial age, where everyone had to be at the conveyor belt at the same time for the production process to work.
With the rise in knowledge work, that requirement is going away in many places. It makes more sense to ask what the best work schedule for a knowledge worker would be for them to achieve peak performance. For some it might be a three-day work week; for others it might be a different arrangement.
Perhaps we even must reframe it from “work time” to “time when you are contactable” and from asking people to sell you their time to selling you their productive output. You see, when you get your best work ideas in the shower, the concept of a workweek goes out the window!
Alexander: Earlier you mentioned self-driving cars and digital patient records. In many countries, efforts to further drive these innovations are hampered by local regulations. What role does the government play here?
Florian: If we, as a society, are to benefit from all the new technologies that are becoming available, regulation needs to not only catch up but actually spearhead the development, especially when it comes to crucial infrastructure investments.
In terms of self-driving cars, a lot of the public discourse centers around the question of who is liable when something goes wrong. That is an important question, but it shouldn't dominate the discussion and hamper any progress. This is a technology that already can save lives and will do so even more in the future. So, any day that goes by that we don't make progress on this front means more people will unnecessarily die or get injured in road accidents.
When people rightly point out that the computer vision used in self-driving cars sometimes struggles with bad road markings, that shouldn't be a warning about how bad self-driving cars still are. Instead, we should go out and make sure lane markers are painted properly. That is a paltry cost compared to the huge benefit we would derive from it.
You could go further: Why do cars have to learn how to “see” lanes and stop signs and so on in the first place? Why haven't we thought about using technology so that the infrastructure can communicate with cars more directly? It is not that far-fetched when you think about the fact that trains receive signals from the tracks that they need to stop automatically when there is a red light, or that airplanes receive radio signals from the runway to indicate the perfect glide path. Incidentally, trains and planes are also much safer, generally speaking, than cars.
It is a similar story with digital patient records. I see, of course, the sensitive nature of digitizing this kind of data, but it is a tractable problem that can be solved, and, again, the opportunity cost of sitting on the problem is bigger than doing something about it. Countries like Canada, Denmark, and Estonia have shown that it can be implemented safely for the benefit of society. Other countries can learn from them.
Alexander: Thank you, Florian. What quick-win advice would you give that is easy for many companies to apply within their digital strategies?
Florian: Find out what the key strengths of your employees are! We have talked a lot about new digital tools, but who are the right people to operate these tools? There are two mistakes that are being made in this context.
The first is to think that just because the tool is simple to operate, anyone can now do so. I have seen this with the commoditization of survey tools that make it easy for anyone to create survey forms. It actually takes quite a bit of knowledge to set up a survey that will yield reliable results! That doesn't mean you have to hire people with the right skills for every task at hand. But you have to ask who might be able to learn the basics of good survey design.
There are of course many tools that don't require any specialized skills. Still, it would be a mistake not to think about who should be operating them. With modern content management systems, for example, it is easy to update the content of a website.
