Big Data. Data Analytics. Machine Learning. The power of data. Analytics providing understanding and advantage. The phrases fly thick and fast, and seem to be everywhere, with every firm wanting to exploit the power of data and data analytics to gain an advantage. One memorable quote states that, “Without big data analytics, companies are blind and deaf, wandering out onto the web like deer on a freeway” (Geoffrey Moore, author and consultant.)
But what exactly is data analytics and what types of data analytics are there?
Data analytics in brief
Investopedia defines data analytics as ‘the science of analyzing raw data in order to make conclusions about that information… [it] is a broad term that encompasses many diverse types of data analysis. Any type of information can be subjected to data analytics techniques to get insight that can be used to improve things’. So far, so logical. In essence, data analytics could be viewed as a fancy name for something that people have been doing since the a Neanderthal first tried striking rocks together over some kindling to start a fire and figured out that using certain types of rocks and kindling that was drier made the whole process more successful.
There are, however, several factors that make data analytics today qualitatively and quantitatively different from that previously:
- Firstly, the level of automation that is used. Data analytics techniques and processes (as well as those for data collection) have increasingly been automated, which has increased almost exponentially the amount of data that can be analyzed.
- Secondly, data analytics has become much more available and accessible. It wasn’t too long ago that data analytics was something that only governments and large corporations were able to do (and even then, it was only certain parts of governments and corporations!) Now, almost anyone with an internet connection can get access to advanced data analysis tools and methods.
- Thirdly, the various types of data analytics have become more recognizable and easier to differentiate for users.
Types of data analytics
Data analytics is broadly broken down into four basic types:
- Descriptive analytics: this examines data for information, patterns and detail as to what has happened over a particular period. The information examined varies between industry, sectors and companies but could answer questions such as ‘how have sales changed in the last year?’, ‘has the number of listeners / views changed?’, ‘how many medicines were issues last quarter?’ and so on.
- Diagnostic analytics: this examines data for the answers to the why questions and is more complex, involving the generation and testing of hypothesis. ‘Sales have gone up – why? Did the opening of new stores have an impact? Did a marketing push have an impact? Was it the result of a new affiliate?’ Or, ‘admittances to this hospital department have risen – why? Are more people ill? Or is it the result of a better awareness campaign? Or because a hospital two counties away has closed a ward?’
- Predictive analytics: this examines what is likely (or unlikely) to happen in the future. We have all become familiar with the phrase that ‘past performance is no guarantee of future results’. Predictive analytics seeks to examine a range of data to determine the future course of a particular area of interest and in a manner that is much more reliable and sophisticated than a straight-line extrapolation from the past and which aims to avoid human cognitive biases.
- Prescriptive analytics: this type of analytics seeks to recommend or suggest a potential course (or courses) of action. This forms the basis of many quality control systems that examine large volumes of data and recommend measures. Predictive maintenance, for example, is one of the most interesting and valuable areas where data analytics is making an impact, through examining data, spotting patterns for when failures occur, and then issuing alerts ahead of such failures occurring again.
Data analytics daily
These four types of data analytics underpin many of the activities that take place each and every day. It does not matter whether it is customer analytics for a retail company, advanced data analytics for a industrial conglomerate, or more efficient data collection and analysis for policy makers, data analysis is a fundamental part of all of our daily activities. With the increase in the availability of data, the analytical methodologies and tools, more and more companies and organizations are using data analysis to help with their particular activities, to give themselves an advantage, and to make their actions more efficient.