What is Analytics?
That’s one of those questions that should have a simple and straightforward answer, but in fact has dozens of answers (not all of which are either simple or straightforward.)
Addressing a closely related question, Thomas Wailgum at CIO is wondering just exactly what we mean by the term “BI.” These two questions are linked because the two areas of expertise are closely linked, and because another question that often gets raised is “What is the difference between BI and analytics?” (or vice-versa.)
In fact, Tom Davenport, one of Wailgum’s panel of experts (along with Merv Adrian and others) offers the following:
I view BI as all the things that organizations do to make sense of data for purposes of managing an organization better. To me it includes reporting and analytics, though there is much more of the former than the latter. However, I sense that the BI term is rapidly being supplanted by “business analytics.”
Tom is absolutely correct in setting out analytics as a subset of Business Intelligence. Analytics is one of the “things organizations do” to transform data into effective business decisions, but it isn’t the only one. At Sybase, when distinguishing between the two categories that we have labeled Reporting and Advanced Analytics, we first look at whether data is simply being formatted and summarized, or whether deeper patterns within the data are being sought after and explored. Advanced analytics generally requires extensive ad-hoc analysis as well as consideration of many more variables than are likely to be incorporated in a simple report. The goal of an advanced analytics environment is often forecasting — providing a sufficiently detailed analysis of what has come before in order to make plausible predictions of what will come next. The term “predictive analytics” is often used to describe such applications, but it’s important to note that any sufficiently complex analysis can be considered advanced analytics even if it does not provide a prediction or forecast.
For example, any time that human behavior is being analyzed, that’s most likely advanced analytics. To be sure, businesses often analyze behavior in order to make a prediction about an individual will do next — Buy more of the same? Buy something else? Default on payment? — but they can also use behavioral analyses to determine how effectively their systems and processes meet customer expectations, how well they have done at holding the target market’s attention, and related questions.
The vital distinction is this: advanced analytics involves more than just slicing and dicing of the data. The data has to be evaluated for key trends and patterns, which are abstracted into a model. When new data is gathered it is run against the model. To give a very simple example, analysis of customer purchasing trends may reveal correlations between a customer’s past buying behavior, geographic location, income, and likelihood to purchase one of several new products. When a customer carries out a transaction, his or her behavior and profile are compared with, or scored against, the model in order to predict the likelihood that a particular offer will succeed.
Ultimately, it’s this reliance on models that sets advanced analytics apart from other types of BI analysis. When a business takes a look at data to try to improve decisions and performance, that’s business intelligence. When a business compares incoming data with a model in order to achieve deeper understanding, deal with human behavior in real time, or predict what’s going to happen next, that’s advanced analytics.