Over the years, most organizations have focused their attention on the effectiveness and efficiency of separate planning functions. As a new way of doing business, a growing number of organizations have begun to realize the strategic importance of driver based planning powered by SAP Business Planning and Consolidation (BPC) on SAP HANA. SAP BusinessObjects Predictive Analytics (PA) supports an algorithmic approach to forecasting future values in a time series, while at the same time identifying the business drivers. It uses state-of-the-art Machine Learning algorithms.
In an effort to help organizations capture the synergy between PA and BPC, this blog will provide information on how to make better data-driven planning decisions leading to increased operational efficiency by describing how:
- PA forecasts future values and identifies the business drivers
- SAP Design Studio application enables the business user to simulate forecasting various scenarios running PA’s Machine Learning algorithms in the background
- Predictive forecasts can be written back to BPC on SAP HANA
The current blog explains the conceptual process and the second part will focus on the technical implementation.
First, let us explore what predictive forecasting is and how it combines statistical analysis and key trends as they happen to predict results in the future.
The predictive insights enable companies to maximize profitability and operational efficiency when combined with BPC.
Over the years, enterprises have collected data from divergent systems and have been aggregating financial planning information with tools like BPC and offline they are building up drivers. However, it is evident that most of the times the drivers are estimated to their best of knowledge and not based on the patterns found in the data.
The driver based plans are not effective unless you start with accurate inputs as planning accuracy is heavily dependent on the drivers and inputs used in the planning process. This gives life to the idea of combining predictive power to financial forecasting. PA has the ability to predict more accurate inputs for planning systems.
So far we have comprehended that combining the PA and BPC increases the planning accuracy. Now in the real world, inputs such as sales volumes, raw material prices, etc. have a significant impact on forecasted prices and profitability. These inputs constantly change over the time period. This necessitates the ability of Financial Analysts to respond to the changing market dynamics.
Using SAP Design Studio, enterprises can build effective interactive analytical applications targeted towards these Financial Analysts. A solution was built by integrating SAP Design Studio, PA and BPC to provide the ability to simulate the drivers, to predict the forecast using PA’s Automated Analytics Time Series Forecasting and to save these forecasts to BPC.
You may watch the recorded webinar by Jon Essig from SimpleFi Solutions, who elaborated a scenario based planning on cotton price forecasting with multiple drivers. He also explains how a Financial Analyst can simulate the drivers and forecast using the PA and eventually persists the predictive forecast to BPC.
To realize the benefits of PA with BPC, one does not need to possess data scientist skills. Financial Analysts can simply build necessary PA models to identify the drivers, do forecast planning and hand over the baton to BPC.
Please watch out for the next blog which explains how this solution was implemented technically.