Big Data Analytics

Forecasting the sale of ice cream in the supermarket based on weather forecasts and historical sales data; monitoring the spread of diseases based on search volumes; predicting the need for repair of aircraft engines in real time – the possible uses and potential of big data analytics are undeniably inspiring.

Figure 1: Overview of typical applications of big data

But what is really behind this alleged “crystal ball”? Big data analytics is the study of large quantities of data using various statistical analysis methods. The aim is to identify hidden systematic patterns, correlations and interrelations from which decision-relevant information, optimisation potential and – where possible – sustained competitive advantages can be obtained.

Figure 2: Big data analytics funnel – from big data to meaningful results

As can be seen from Figure 2, big data input can consist of various data forms and can come from different sources, e.g. sensors, articles and more. What makes big data challenging from a data management perspective are the following three factors:

• Volumes: The volume of big data exceeds “normal” dimensions, usually amounting to several terabytes; it can even push into the petabyte range, depending on the data at hand.

• Diversity: Big data encompasses a wide variety of data types – it consists of both structured and unstructured data and data sources.

• Speed: A further defining characteristic of big data is the speed at which it is generated – the frequency of updates through – for example – real-time data, automatic recording systems, etc.

As part of the analysis phase, the data sets are compressed in volume by means of various processes and their quality enhanced. More complex analyses, such as predictive analytics or data mining, often make use of purpose-designed software tools. The foremost objective of a comprehensive big data strategy is to use big data analytics results to help a company …

• …make quicker, better decisions…

• …achieve cost reductions based on novel findings (e.g. the scheduling of repairs in advance of machine breakdowns) …

• …develop new products and services and, where necessary, even new business models (e.g. as a result of findings on customer behaviour, etc.)

As part of this process, we can help you identify and examine the aspects of big data analytics that are of strategic relevance for your organisation and then prioritise them according to their value contribution. Thanks to our involvement in numerous projects across a variety of sectors, we draw on plentiful experience in the development of value-creating strategies. Please don’t hesitate to get in touch!