You are probably familiar with them, those technology trends that emerge and you suddenly hear all around you. Within the field of Business Intelligence (BI) we are familiar with them, too. One of the trends that BI companies are obsessed with is Predictive Analysis.
Simply put, Predictive Analysis is predicting with data what will happen in the future. In technical terms, this means extracting information from existing datasets with the help of predictive models and statistical analyses. This allows patterns to be found in historical and transactional data. These patterns can then be used to make decisions aimed at managing the business (more) effectively.
Back to the question in the title. Is Predictive Analysis simply hype or genuinely the future? A breakthrough or the same little sweet in a new wrapper?
Predictive Analysis is a lasting trend. A development that meets the needs that many companies are struggling with. You see, it is a logical continuation of the development introduced in recent years in the field of data analysis and BI.
The model below from Gartner provides a good illustration of this development. With the help of Descriptiveand Diagnostic Analysis, companies link the data available in the company. This is the simplest form of analysis. Just think of a statement showing the movement in revenues, margins, customer purchases and hours spent over time. Predictive Analysis goes one step further. Predictions are made about the future based on the past. Then, in turn, Prescriptive Analysis builds on this, answering the question of how we can respond to the predictions made.
All well and good, you think, but isn’t this typically something for nerds and geeks? What can I do with this in practice, and how will Prescriptive Analysis help me to understand my customer? To explain that, I recommend you read this blog by Jake Sorofman in the Harvard Business Review (HBR).
The article starts with a concise description of the new consumer: hovering in the shop, bent over his or her smartphone busily comparing prices and reviews before he or she makes a purchase decision.
Applications that offer support
Recognise this? We are in fact opting more and more to consult all sorts of applications that offer us support in making the right choice while we’re shopping. One example is the popular app for wine lovers: Vivino. After scanning the barcode on the bottle this app shows the average price, drinking advice and other people’s opinions. And, well, who wouldn’t want to know what the price was of the bottle of wine you received at your last birthday party?
Jake Sorofman describes in his blog the development of consumer behaviour and strategies for responding to this. In the blog he underlines the importance of data and how to use it to present every customer with a personalised offer. Predictive Analytics make this possible.
Meaningful use of data
Companies generally know that there must be a lot of information from customers ‘somewhere’ in their systems, but to make meaningful use of this data is extremely challenging. The trick is to turn the complexity ‘under the bonnet’ into a user-friendly app which the business owner can use to provide a personalised service to his customers and which, at the same time, can make the customer feel that he or she enjoys preferential status.
Finally: The success of Predictive Analytics depends – as always in IT – not on the tool used (the data or technology). It is dictated by the extent to which companies are able to apply data as a strategic business asset. This therefore starts with asking the right questions. What do we want to base our decisions on? What is the objective (reducing costs, maximising profits or achieving customer satisfaction)? How do we want to achieve that?
Only after that do we look at the available data which will form the basis of the analysis. The right knowledge and tools are then needed to be able to conduct the predictive analysis.