Which companies have big data

Why big data is often useless

The fact that this whole big data story is not yet running as smoothly as the theory and the providers who make good money from the hype promise it was again made clear at the beginning of February by a simple and intelligent question.

It was presented by Helmut Krcmar at the Hamburg IT Strategy Days. The professor for business informatics at the Technical University of Munich will moderate the event together with CIO editor-in-chief Horst Ellermann. The question was: "Why do I see advertising on the Internet for weeks after buying a trolley case?"

Presumably almost everyone in the audience had asked this question before. Accordingly, they were excited about the answer.

Which then turned out to be less than satisfactory: "The technologies are getting better and better, but we are still in the early stages", said Daniel Keller, CIO of Axel Springer Verlag.

Much is possible without big data

Targeting, recording, storing and sorting the traces that millions of customers and potential customers leave behind on their journey through the Internet is a classic big data application. The principle itself is older than the term.

"To be still in its infancy" means less that the makers only started last week. Rather, it is said that even after years of efforts, data-driven business models do not work as easily as a cigarette machine, into which we insert the appropriate money at the top and exactly the desired result - i.e. the right pack - comes out below.

The thing with the rolling suitcase is probably due to the "algorithm weakness" that is widespread among machines, i.e. the inability to draw the right conclusions from a lot of collected information.

That being said, there are two main reasons why companies do not benefit, or not enough, from big data. The first: With the help of data analysis, you get results that you could have had with data that was not quite as "big".

Much cannot be implemented in practice

In this context, the Harvard Business Review reports on a financial services provider who, with the help of large amounts of data, developed models that can be used to determine the best location for installing ATMs. After they were done with it, those responsible stated through a note that such models have been around for years ...

The second: Big Data produces results and ideas that, for whatever reason, cannot be implemented in practice. A large US retailer had found in a model test that sales increase if you put a special offer product on the shelves a while before it is cheaper and leave it there when the offer price is no longer valid.

In order to implement this principle in all branches, the company would have had to turn its entire supply chain upside down. You didn't want to or couldn't. So in the end the data analysis was useless at this point.

  1. Scouting with Scout7
    Football clubs no longer leave anything to chance when searching for talent: Specialized data analysis tools such as Scout7 record and evaluate around 3000 games every month.
  2. Player profile
    For every player there is a profile page in the system with extensive statistics.
  3. Video clips
    Decisive moves can be called up via video clip.
  4. Performance data
    The movement of each player is precisely recorded in every game.
  5. Club management
    SAP offers a comprehensive management solution for sports clubs based on its Hana platform.
  6. Training sessions in the film
    During training and games, cameras are used to record data on the basis of high-resolution videos.
  7. All-round solution
    The sports business today is a complex relationship between fan, club and players. Real-time analysis of data enables clubs to make well-founded decisions in many areas.
  8. Even in the NBA
    SAP's big data systems are also already being used in other commercial sports such as the American NBA (basketball) and NHL (ice hockey).

It is not about evaluating data as laboriously and extensively as possible and then thinking about what you want to do with it, but rather using existing knowledge instead of gut feeling as the basis for decisions.

For laundry, sweet potatoes and peanuts

This knowledge has to be "big", the data does not necessarily have to be. On the contrary. David Meer, partner at strategy & (formerly Booz & Company), thinks that "little data is also important."