Digitalisation - change the right thing with the right data

Find the root cause with digital support

In today's organizations priority is often given to traditional KPIs that indicate the status of time, quality and finances. These KPIs are also called "lagging indicators", and are not considered motivating for the individual. Therefore, Terese Janson, CEO of Prindit, argues that the metrics systems should also be supplemented with "leading indicators" that focus more on how the employee feels and wants to perform.

However, a challenge is that there seem to be a general feeling of having a good grasp of your organization, which according to Terese is often wrong as you tend to confuse symptoms with the root cause. There is a so-called "Perceived truth" for each individual who tends to be linked to the symptoms, which in turn result in leaders having difficulty navigating and thus acting on symptoms instead of the root cause.

One way to solve this is to see each employee as a sensor where you can collect a lot of data about how they feel. This leads to a more objective picture of the current situation, which is also presented in real time. Through all the data, you can see the status of each individual, group and department, as well as of the organization as a whole. By analyzing the data it becomes easier to see what is actually the root cause of a problem, and you can then act on that instead of the symptoms that appear in the organization. It will also be easier to see what measures are needed in which part of the organization, instead of trying to find a general medicine that works everywhere. In addition, this data provides the ability to see trends and analyze the effect of different measures. By having the right data, one can thus understand the root cause of a problem, and then act on the right basis.

The webinar was arranged together with Underhållsmässan.

Get to know your target audience

Another example that illustrates the importance of the right data is from the automotive industry, where the background problem is the need to recruit more drivers in the future, but where current recruiting tends to only focus on half the population, men. With this in mind, Tatiana Temm, Communications Strategist Ampersand AB, share insights from a project where they looked at why there is resistance among women to work as a truck driver, and what can be done to get around this problem. The study examined physical differences between women and men (they all were 170cm tall), and compared this with the design of a truck cab. The study showed that the driver environment is much less adapted for women, e.g. due to shorter range which makes it difficult to reach instruments. According to Tatiana, the solution to this problem is to get to know your target group, that is, to ask the right questions in order to get the right data. One methodology suitable for this purpose is the LASS methodology (Look - where growth is / where new money comes from, Ask - what are they annoyed about today? / What is the problem, Solve - solve the problem, Sell - take the solutions to the market). Based on the "correct" data that comes from applying this methodology, you will probably need to work with one or more of the four following innovation areas to get the desired result, in this case be able to recruit female truck drivers.

• New / modified product
• New message
• New channels
• New sales methods

AI and predictive maintenance

A third example of how the right data can guide you into making the right decisions is picked up from the field of maintenance. Jeremy Jean-Jean and Wilhelm Vermelin, researchers at RISE, point out that there are two major challenges in maintenance today; (1) to know how often maintenance should take place, but also (2) to understand how the component/machine can be fully utilized before maintenance.

To find the most optimal strategy for maintenance, one should strive for data-driven decisions, rather than experience-based decisions that are often the case today. One way to succeed in this, and thus be able to achieve predictive maintenance, is through machine learning and AI.

In practice, this means that data must first be collected, for example through sensors on the machine that sense and register vibrations. Patterns in this data can then reveal when maintenance is needed (machine learning is algorithms that detect these patterns). However, the prerequisite for success is having the right data – available and reliable.

Panel Discussion