Business Intelligence
- Details
- Created on 20 October 2008
- Written by Steve Burrows
What is this software that the IT industry calls "Business Intelligence"?
Elsewhere I've tried to explain Data vs. Information, and the concepts of Information Systems and Business Systems, where data is converted into information and used in the context of a business process. However these transactional systems deal mainly with specific items of data - this customer, that order etc., in the execution of a business process.
As business managers we have a need for higher level information based upon aggregated data - e.g. total value of sales for each salesperson last month, number of prospects in Leeds vs. number of prospects in Manchester, and so on.
We use this information to manage performance, foresee trends etc.. A lot of it is quite complex to generate, there are many conditional factors to consider, and IT people use a set of specialist tools to assist with generating easy to digest information out of complex and conditional data - these tools are generically described as "Business Intelligence" tools. The name is meaningless - in the past they have been called "Management Information" tools, and simply "reporting" tools, but their function remains relatively constant - to collate and transform large quantities of data into high-level information - reporting by another name.Components of Business Intelligence suites typically include
- Reporting: the presentation of aggregated data as a document of information - a "report"
- Analytics: extraction of filtered data into a tool such as a spreadsheet so that the user can experiment with looking at the data in different ways
- Dashboarding; presentation of aggregated data as simple key performance indicators (KPIs), commonly in a graphics form (barcharts, speedometer displays etc.) and automatically updated very frequently(every few minutes) or in real time.
- Data Mining: tools for examining data to identify characteristic patterns / repetitive patterns / trends.
- Data Integration / Extract, Transform & Load (ETL): tools to extract data from various data sources, transform that data into something more meaningful, possibly combining several data sources together, consolidating data etc., and then reloading it in its new enhanced form for mining or reporting.
- Web Portal: a server side tool allowing reports and dashboards to be viewed via a web browser without access to any special Business Intelligence software on the local computer.
So "Business Intelligence" is conceptually pretty straightforward stuff that has been linguistically dressed up by vendors to increased its percieved value. Get data, mash and manipulate it into a form whereby it becomes meaningful, summarise it and report it. While it sounds simple it isn't. Typically the data comes from multiple data sources which may need integrating together, the volumes involved may be huge - it is perfectly reasonable to be reporting on trends extracted from the last million events, and these potentially complex and large manipulations are expected to be performed in near real time. Which is why the vendors of these reporting / information extraction technologies dress it up - they are trying to create a sense of value for what many users and information recipients see as a simple task, but which is in reality often very, very complex, but gives us high value information about performances, trends and behaviours within our businesses - hence the name Business Intelligence.