Starting a pool of data scientists within innogy that serves its own business departments. That was the plan, says Robert Egging, International Head of Retail Data Strategy at innogy (E.ON group). Until they investigated on which themes the different departments wanted data support. The need turned out to be on another level.

Analysis capacity was no problem, Egging’s team discovered thanks to a so-called ‘data stock taking’. “Our people were able to build the most beautiful models from a multitude of sources. But they gave limited answers to the real needs of business owners. As a result, the efforts in the field of artificial intelligence and data science yielded less business value than could potentially be achieved.”

Execution power

There was a particular need for an ‘execution manager’: someone who ensures that there are models that do fit seamlessly with the user stories of the business owners. A person who also involves the controller, so that he also gives his blessing to the model and thus signs for the expected value of the requested investments. “You can’t leave that validation game to every data scientist. Most of them are very unhappy about that,” Egging said.

Empathy

And so he set out to find what he jokingly calls a “consultant and nerd all in one.” He couldn’t find someone like that internally. Fortunately, Pipple turned out to be a large data scientist with great empathy. “Business savvy mathematicians work there. They like to make a model. But also to sit at the table with a management team. And to explain in simple terms to teams what they are doing and what the value of that is.”

Scale up

The value of working with Pipple is clearly visible. For example, they can now predict which of the large business customers can easily pay their future energy bill, which leads generate the most conversion and there will be models that help determine which offer best seduces which consumer.
Egging sees the opportunities only growing further. “We have 22 million innogy customers. The impact of a working model is quite quickly large, because it can be scaled up almost endlessly: you can apply the same trick to even more segments in even more countries.” This is reinforced by the acquisition by E.ON. “That means more than doubling our customers. And our new colleagues also have all kinds of models on the shelf. By sharing our knowledge, we can make huge strides together.”