T he future of a digital economy depends largely on the availability, analysis and interpretation of data. Already today, countless amounts of data are generated and processed for various applications. In the field of marketing, we are confronted with this every day, from transactions with credit cards to the systematic recording of surfing behavior on the Internet. Data collectors everywhere are waiting to document our behavior and forecast conclusions for the future. The vision is that at some point algorithms will be able to predict our behavior so accurately that we will always get exactly what we are interested in or need at that moment.
In addition to human user data, data is playing an increasingly important role in the industrial context, for example in the control and optimization of manufacturing processes. With the help of predicted user and machine data, it will be possible in future to link all these data streams with each other and thus a presumed change in consumer interest will lead directly to a control of production in the company.
Everything is networked with everything else, and the data streams play on this network. Regardless of whether it is a question of forward-looking maintenance or expected consumer behaviour, the aim is always to use data to make reliable forecasts about the next steps, the upcoming development or even possible events. Every company will have to deal with this, be it in addressing customers or organizing production processes.
The example of customer-specific advertising illustrates the task that the algorithms play in processing. They bring together information from a wide variety of sources, structure the information components and, on this basis, develop a customer-specific offer:
It is understandable that the amounts of data generated here can no longer be evaluated by one person alone. This requires computers that can search enormous amounts of data for correlations at enormous speed. A further advantage is that these algorithms can also identify interesting connections in unstructured data under the assumption of unlimited computing capacity (the so-called “big data approach”).
For a better understanding of the significance of the data, a comparison with the combustion engine is used. Depending on design, this engine has a smaller or larger capacity (performance). However, it can only develop its power if it is supplied with fuel and thus starts running. It is similar with the digital infrastructure. All sensors, computers and output devices can only develop their added value if they collect and evaluate data and deliver results.
For the companies this means that they should intensively deal with data acquisition and data evaluation in order not to lose touch with the digital world. This should go in all directions, from the internal production processes to possible data of the end users of their products. The key questions here are: What data do I already have in the company? Which data can I get from suppliers or customers? And how can I generate or organize missing data?
Companies often do not have to create new data sources at all, as there is a lot of information available. Modern machine tools, for example, generate extensive data records on the work process and its condition for every customer. At the same time, data can also be shared within a value chain, which can be useful in many applications to create transparency in real time.
However, in order not to be lost in the flood of data, companies should first consider which data can be sensibly linked with each other. Otherwise you run the risk of losing the overview and leaving a lot of data unused.
Basically, we distinguish between three possible cases:
While data for (1) can be easily processed using existing software, greater strategic effort is required for (2) and (3). Only having the data is no added value for the company, only if the findings are used for process optimizations or new business models, a real benefit for the company arises.
Particularly in Germany, companies are confronted with strict data protection, which significantly restricts the imagination, especially for the end user. However, companies should not be discouraged from leaving their data unused. In the meantime, for example, there are competence centres where companies can experiment with ideas for new business models.
It is important to discuss rights and possible legal framework conditions with the relevant experts at an early stage. It may, however, be worthwhile advocating exceptions to the otherwise strict rules, especially in the area of digital regulation there is more room for manoeuvre than in the established areas.
Working with data is not only interesting, it can also lead to surprising findings. A lot of information is already recorded somewhere in the companies, but is not evaluated. For the start into the data world of tomorrow, the following steps can be meaningful: