There is a wide range of analytics tools that can be used to conduct analysis of large sets of data and transactional activity. Analytics has increased in popularity in the past few years and is forecast to continue to experience above-average growth well into the next decade. As technology has improved and received greater acceptance, organizations have been collecting and storing large quantities of transactional data. The purpose of analytics tools is to use this data to determine patterns and trends. This information can be used to assist in the decision making process.
Analytics tools can be divided into three categories: interest indicators, activity evaluation, and data selection. Although many people assume that analytics tools are a new development, they actually represent some of the oldest concepts in statistics and data management. The advent of the Internet and the desire by businesses to track the effectiveness of this tool in reaching clients has fueled the rapid growth of analytics tools. In order for any organization to determine how many resources to allocate to the Internet, metrics are required to determine the return on investment and the relative usefulness of this tool.
Interest indicators are the most common of all the web-based analytics tools. A small program or script is added to the website to track user activity. The most basic tools can provide a summary of user country of origin, time accessed, browser used, total amount of time spent on the website, and reference source. More complex, commercial products can provide the exact Internet protocol (IP) address, the number of times the same person has visited the site within a specific time frame, where they went, and how long they spent on each page.
Activity evaluation tools can range from simple data collection to an evaluation of the business process. For example, a web-based tool can provide a summary of the most common access paths, time spent at each stage, and users who have accessed each data table. For a transactional system, the same type of analysis can be completed using a combination of information from several tables and databases. The tools used for this type of analysis are typically quite resource-intensive, requiring significant hardware and storage space to operate.
The data selection or data extraction tool kit is used to move specifically identified data from the transactional database into the data analysis warehouse or cube. The specifications must be quite exact to build the appropriate data set for the analysis tool to use. Too much data is costly, and not enough data will not provide accurate results.
