In life, we rarely turn around to look behind us as we keep our sights focused on what’s in front of us. So in analytics, why are we focusing so much attention 
on what has already happened, without applying it to what could happen? The goal of predictive analytics is to answer the what could and what will – two things all companies want and need to know.
In a recent study, Gartner forecasted that 50 percent of all employees will need access to analytics by 2014, but today, fewer than 10 percent of employees have such access. Employees are lacking access to historical analytics, real-time analytics, and more importantly, predictive analytics, which are key to decision-making. This disconnect demonstrates the need for a shift in strategy and focus for companies who are looking to remain competitive and sustain an edge.
This shift from just making decisions to following the data to decision path and making all the right moves along the way include being informed and using analytics, from the most basic, to the most complicated.
At this point, most companies using data and analytics understand the type of data they need to collect, best practices on how to collect, and they know where and how to store it securely. The next phase gets into how the data and analytics can be processed and used, which includes a heavy focus on anticipating the future and using predictive analytics.
Predictive analytics refers to the process of reviewing historical data and applying it to future situations based on conditions, and creating predictions for performance. Predictive analytics is a combined art and science that processes data in a typical way of data mining data sets, then adds a layer of analysis based on past experiences and qualitative factors to yield predictions.
Predictive models, often created by data analysts, exploit specific data patterns and identify risks and opportunities. The models do this by capturing relational information among factors and encouraging assessment of risk or risk potential associated with the set of conditions. These models help with decision-making and invite forward-thinking decision-making rather than making decisions based solely on historical data sets.
For a highly seasonal high-tech company that has trouble understanding the impact of cash flow and was experiencing delays between bookings and billings, they realized that they needed to accurately forecast revenues and margins to identify underperforming parts of the business. Through the direction from SAP’s Performance and Insight Optimization (PIO) group, the company realized that they needed to take advantage of each supplier’s/buyer’s unique seasonal behavior and use the HANA database for recognizing growth rates and performance differences by region. By understanding these aspects of the business, the high-tech company was able to predict seasonality factors for a trend analysis, clustering of sales nodes (performance analysis), and clustering of products. This data allowed the company to better respond to customers and deliver tangible business value; reduce time to transfer information into insights and improve the quality of decision-making on those insights; and drive higher profitability and growth.
By getting the most of predictive analytic software and creating predictive models and easily pulling from data sets, companies can effectively process the information, apply it, and it can lead to accurately forecast demand and helping a company mindset evolve their focus from sense-and-demand to predict-and-act capabilities. By using predictive analytics in real-time, demand, supply and behaviors can instantly be better managed.
It all boils down to having the right information, at the right time, with the right experience to understand how to turn the data into a decision. Find out how SAP can help analyze and leverage data with the Performance and Insight Optimization (PIO) group.
