Nowadays, data is often referred to as 21st century oil – a vital resource for any business. So how has handling data changed over the last decade?
MICHAEL FEINDT Companies have always collected information about their customers and their own financial situation, but the way in which it is stored has fundamentally changed. Information that was once entered in Excel tables or jotted down on notepads is now saved in databases, which makes data access and analysis much easier. This also has an impact on data quality, which is much better than it was ten years ago.
RAINER HILLEBRAND As the Otto Group is a mail-order company with a remote operating set-up, information has always been extremely important. It starts with simple things such as who is calling and what is the delivery address, and continues right up to after-sale customer feedback. Thanks to advanced digitisation, data can be collected using entirely different methods and in greater quantities than ever before. It can also be processed much faster and stored at a fraction of the cost – something we hardly even dreamt about ten years ago.
Large quantities of data are one thing but analysing this is another. What challenges do you face in this respect?
RH The key question is how can we filter out the right information to satisfy customers’ requirements from the mass of data we accumulate? When a customer is looking for a specific product, we have to come up with an ideal, tailor-made solution.
This is where you have your finger on the pulse, Mr. Feindt. With your colleagues at Blue Yonder, you have developed an algorithm that facilitates data analysis and works independently.
MF Our algorithm calculates how our customers’ purchasing habits change on a day-to-day basis. This is very important in our customer approach, as Mr. Hillebrand has already pointed out, but it is also important from a business management perspective. For a mail-order company, inventory management and speed of delivery are crucial. These processes are optimised in our case through artificial intelligence.
How does this work?
MF As much information as possible is analysed focusing on historical sales data, prices, discounts, the weather forecast for the next few days and the season, i.e. any factors that can influence a purchase. The system does this for every item in the range. This helps the retailer to decide which products will be ordered from the manufacturer and in what quantities, and which products are not so urgent. The goods are already in the warehouse before the customer orders them, which obviously speeds up delivery times. Furthermore, you store as many items as you need. This means that no bulk quantities are left over at the end of the season and, from a supermarket perspective, there is far less waste.
Your artificial intelligence system works independently and automatically. It takes on tasks previously carried out by people because it is faster and more efficient. What does this mean for the corporate culture of the Otto Group?
RH It means a change, of course, but people shouldn’t view the use of algorithms solely as a threat to jobs. Algorithms help us to improve our customer service and operate more efficiently. That’s good for the company and therefore for every employee. We have to face the challenges of the future. The Otto Group is a large company with numerous options. If we lose one task today, a new one is being created elsewhere in the company. Obviously, employees have to be more proactive today. It’s no longer a case of learning something and resting on our laurels for the next 50 years. Job descriptions are constantly changing and people have to retrain and learn more. But we are helping our employees to move with the times.
MF Digitisation is an innovative process as we have seen over and over again in the history of mankind. CERN, for instance, the European Organisation for Nuclear Research, where I worked as a physicist, was created sixty years ago. At that time, hundreds of women sat in huge halls with slide rules and logarithm tables, and calculated what physicists needed for their work. These jobs no longer exist having been taken over by computers. But would we want to go back to those days? We are currently in a transition phase. Throughout history, the more efficient has always prevailed so we certainly cannot stop digitisation. It’s definitely not a case of replacing people but about helping them to improve I’m convinced that the relationship between man and machine will become closer because human beings can do a lot of things beyond the realms of artificial intelligence.
MF A machine cannot think strategically about formulating corporate goals for the coming years. Similarly, it is light years away from designing next season’s fashions. Mankind is always superior when it comes to creativity. But machines optimise processes more efficiently, e.g. repetitive processes such as the ordering of goods. Let’s consider the partnership in this way: man devises the strategy and the machine implements it. I strongly believe that there will be as much work as there is today. Fact is that at some point we just did not need a boiler man on trains anymore either. But people don’t have to do all the donkey work.
RH It’s like that: People have a lot of experience. And experience is based on information we accumulate through life. Algorithms can connect this wide range of experience-related data in a more logical, stringent, error-free manner than we can. But they need processes within a narrow framework. When we make a decision, lots of things come into play – a certain knowledge, experience and gut feeling that software simply doesn’t have – at least, not yet.
Today, people are worried about their jobs on the one hand and their data on the other hand, which is collected by many companies. How do you allay your customers’ fears?
RH On the one hand, we follow the legal framework rigidly. The Otto Group is a family-run business and we rely on our good, reliable reputation built up over many years through a correct approach, values and social responsibility. It would be foolish to risk this reputation. At the same time we want to offer our customers great service. To do this, we need as much information about our customers as possible. Everyone is familiar with this predicament. On the one hand people are worried about what will happen to their personal data – will it be misused? But on the other hand, everyone wants a good, personal service. We have learned from experience that customers are happy to pass on personal data to us if it means an improvement in the end product and we explain how the data will be used. If we don’t adopt this approach, people are rather reluctant to pass on information.
MF No company can afford to work against its customers and upset them by acting unscrupulously. But something strange is happening with data protection in Europe. I have chatted to younger people about this on numerous occasions. I would say that 95 percent of them are not interested in what companies do with their data. Many do not know that messenger services read every message and create profiles of the sender and recipient and their networks, and sell these to the advertising industry.
But does it really bother them?
MF People say that the younger element doesn’t care. Young people are not bothered by the advertising they receive through social media. They have a relaxed attitude when it comes to their data and US companies in particular make full use of this. We have to ask ourselves if we are placing ourselves at a competitive disadvantage if we are too reserved. I personally think that our attitude isn’t bad but our competitors are far more aggressive than us. How do we deal with this?
We don’t talk about it enough.
RH This is why the Otto Group is constantly calling for equal competition between American and European companies. In addition to the comparatively less stringent data protection in the USA, Amazon, Google, Ebay and even the Chinese companies have another competitive edge. Their markets are huge and they have a lot more customers. They create profiles from the customers’ personal data, link these and deliver the perfect service. Europe is also a big market but the
various languages and jurisdictions complicate matters.
Can Europe still catch up?
MF We still have a chance. Digitisation now ranks highly on the list of priorities of most companies. But change doesn’t happen overnight. Germany is still very conservative, which is perhaps also due to our previous success model. The value of data and software is not apparent here yet. A car has a value that can be grasped. This can also be rather expensive and we are investing billions of euros to develop this. The Americans have long since moved on to this kind of thinking. The world’s most expensive companies are software and data companies. This is where daylight dawns – software is extremely important. It’s a long drawn-out process. I’m not sure if we’ll catch up.
You both work in areas in which forecasts are very important. What will the retail market look like five years from now?
RH I would be happy if I knew what it would look like in a year’s time. Digital change will certainly remain the driving force in retail. The opportunities to collect and store data and make them useable by algorithms will go from strength to strength. After the smartphones success story – 50 percent of our business runs via mobile devices – the next big topic will be conversational commerce and going beyond touch with chatbots and voice input playing a major role in the purchasing process.
What developments lie ahead?
RH Personalisation and individualisation will continue to be key topics along with an even better understanding of what customers are looking for and how they will behave. Blue Yonder will continue to be very helpful for us in this respect. It is fair to say that the progress of the digital revolution is still at its very beginning, basically a few hours after the Big Bang. There’s a long way to go – we haven’t had the full impact yet.
MF That’s true. I think that anyone who hasn’t jumped on the digital train yet will be left standing on the platform. Over the years, several large trading chains have gone into liquidation because they failed to take technological innovation seriously.
And it probably won’t stay that way – right?
MF Anyone who carries on as before and foregoes the ten percent you need to transform yourself with digitisation will disappear. The message is hard but simple. As a company, you have to keep on your toes. New, hitherto unknown technologies will emerge. But the potential must be harnessed and used. What we can predict is that most companies will switch to Algorithmic Merchandising. This means a combination of purchasing, supply chain, advertising, pricing and sales. These elements react and interact with each other and are controlled by algorithms. In the past, different departments did different things – the one hand didn’t know what the other hand was doing. Everything was much less co-ordinated.
And what will the customer’s shopping experience look like?
RH It will look exactly as the customer wants it to!