SAP Predictive Analytics

Intelligent software that helps a lot in helping our various predictions in multiple projects. We do not need to know statistics or special coding to use this platform. We look at forecasting sales and development of our services in different geographical areas. The development and sales department benefits a lot from SAP Predictive Analytics software. When it is possible to anticipate some conditions with software, you can reduce costs and avoid rework. In our work, having information plays a big role in providing advice to customers, which helps us. We only need data, which helps to make more accurate predictions. Data here is like good fuel for moving a car.

When we know that forecasting can do things more accurately and better and predict the [number] of losses and profits, then it can be very useful in work.

SAP Predictive Analytics is used by many teams from the SAP organization which perfectly fits the "SAP runs SAP" mantra. The use cases range from Sales & Marketing to IT operations and events sometimes in the R&D departments. There are plenty of uses cases covered by SAP Predictive Analytics from typical "classification" problems like "predicting which lead is going to convert into an opportunity" or some more advanced like "content recommendation" on the websites.

When you don't have much data science skills or when you need to build something quick, you can get started really fast with SAP Predictive Analytics. There are scenarios, like the ones covered by a neural network, where SAP Predictive Analytics is not well suited, but the big question is usually: "do you really need a neural network to address this use case?"

We have purchased PA but are in the process of proving its worth. It was brought in with the hope of being able to forecast work that is currently done manually, to give an 80% complete starting state that can be altered vs. starting completely from scratch. Additional use cases that will be explored include using historical data to predict trends going forward, which can be used as thresholds to then limit the scope of the investigation. The plan is to use it across our finance organization and even beyond depending on the outcome of our POCs.

It's a great tool to merge actual data analysis (which Lumira doesn't do that well) with visualization (which Lumira does well) - so it can be seen as Lumira for data analysts. However, a lot of the 'predictive' side is hidden/black box which can be frustrating for those analysts, so you could argue it is too complex for casual users, but too 'black box' for analysts.