The field of Analytics is the application of computer technology, operational research, and statistics to solve problems in business, academia, and industry. Analytics is performed within an information system. In the old days, statistics and mathematics were studied without the use of computers and software.

Today, the field of analytics has evolved from the application of computers to the analysis of data and this takes place within an information system or software environment. Mathematics is the core of the the algorithms utilized in analytics. The science of analytics is concerned with the extraction of useful data using computable functions, typically involving the extraction of identifiable properties from large data bases. Analytics combines the fields of statistics, computer science, and mathematics.

Analytics is commonly employed in predictive modeling, in which vast amounts of past data are fed into a computer in an attempt to run an accurate simulation of future events. The problem with predictive modeling is that is is dependent on the variables put into the simulation. There are nearly infinite variables that can affect a simulation. With so many variables, the decision whether to take action on the simulations results is often as risky as taking an uneducated guess.

The findings of analytic analysis that are traditionally the most valued are those classified as actionable, at at least as an actionable insight, An actionable finding is the revelation of a trend or relationship that allows the client to take steps to take advantage of the opportunity.

When a business first begins the run analytics, the exploratory data analysis the initial findings can be very useful. In a nutshell, if an analytics program can answer a business question, such as when the business receives the most orders, which advertising campaign producing a peak in sales, it can be pronounced an early success. Exploratory data analysis can reveal data about a firms clients and suppliers that can be applied to the company’s operations quickly without the need for expensive predictive modeling.

In point of fact though, much of the work done at Analytics companies is not quite actionable. A considerable amount of the daily or regular requirements will be of the “good to know” variety, numbers and data, this is typically referred to as Reporting work. This is the ”unsexy” side of analytics that most analytics professionals try to avoid, favoring the supposedly future revealing capability of predictive modeling. However, just as much useful information can be gleaned from reporting work as from predictive analysis, it just requires a certain amount of manual digging through the data.

The advantage of using analytics this way is that there are insights that can be picked out of the data base that are not as apparent to the computer program as they would be to the human analyst. Taking the extra time to dig through the data and seeking insights that way can save a company from the high cost of building analytic modeling software.

This largely depends on what the company is trying to accomplish. If the client indents to open a new strategic marketing campaign, then perhaps a model is the way to go, as it would allow the client to test many options once the modeling software is in place. If the client merely seeks to understand the drivers of customer acquisition and attrition, an exploratory data analysis would reveal what they seek.

At the end of the day, analytics is a useful tool for companies to understand the drivers that affect their business’s performance, but it is important to remember that the high cost of predictive modeling may not necessarily be justified by the dubious value of its supposedly predictive qualities.

E. J. Peach is an independent blogger and blogs for the erp training site. It’s a non profits blog she uses to present her information to help individuals get ERP certifications and give the newest news about mysql certification questions.