Author: Holger Stelz, Uniserv

Currently, companies are using AI and machine learning (ML) to primarily improve customer loyalty and customer experience. Expectations are raised by ML-based churn prediction solutions to prevent customer churn.

One for all, all for one - this motto of the four musketeers has meanwhile become established in common parlance when it comes to making it clear that "going it alone" is of no use and that success can only be achieved if several components mesh smoothly. This also applies to data quality.

The assumption that forecasts can only be as good as the data on which they are based sounds plausible. As part of a bachelor's thesis at the Stuttgart Media University (HdM), this relationship has now been empirically investigated for the first time and confirmed by means of a concrete test series.