Business Intelligence Architecture

Business intelligence is a fascinating area for many accountants (at least it is for this one). The thing that is most exciting about accounting is being able to provide information that leads to better decision making. Accountants record the historical information of a company. They record sales, expenses, and just about any transaction that takes place. Basically an accountant records the history of the company.

Most history books that you have read are full of the dates of wars and other major events. The historian tries to capture what actually happened (although they are generally biased due to how they view the world). The same is true of accountants. They record the history of the company. Their history books may not be as exciting to read because they are full of numbers (no wars here).

When presidents and leaders of government make important decisions for their country they try to lean on past experience. A president should be well-versed in the history of his country so that he doesn’t make the same mistakes as the people that came before him. Leaders of a business do the same thing. They read through the accounting books of their company and try to avoid making the same mistakes.

Business intelligence is the act of reading through these books and trying to make sense of the information. You can’t read these accounting books from the beginning to the end like you can with a normal history book. You have to filter through all the numbers and find relevant information that will help you make a smart decision. There are a lot of business intelligence software programs out there that will filter through the information for you.

Business Intelligence Architecture

You need to make sure you have the right structure or architecture in place or else the information that your business intelligence system is giving you will be worthless. If you put garbage into the system, then you are going to get garbage out of the system. Every business is different so the business intelligence architecture is going to be different depending on the business. Here is what a normal company’s BI architecture might look like. (see picture below)

Your normal business intelligence architecture consists of several data warehouses (or data marts). For example you might have an accounting datamart, a manufacturing datamart, and or a sales department datamart. The business intelligence program is usually considered the front end. It is the tool that is used to make organize and make sense of all of this information.

Having a solid business intelligence architecture is important. If a building doesn’t have good architecture, then it will eventually collapse. If your business doesn’t have a good BI architecture, then the information you collect will be worthless.

Every company is set up differently. Some companies build products (like Apple builds Iphones and Ipads). Other companies only offer services (doctors and nurses serve patients, but they don’t build a physical product). Because companies are set up so differently, each company is going to have a unique business intelligence architecture.

Standard BI Architecture

Your standard company is going to have a business intelligence program and an operational system. The operational system is just the system that your company is using to gather and store important information. You can think of it as a database of some sort. The business intelligence program is simply the program that your company uses to get information out of the database. Here’s a picture of your standard BI architecture.

Complex BI Architecture

Some larger companies might have more than one operational system. For example, a company might have a database or operational system for all the sales orders, a database for all the invoices, a database for all shipping transactions, and another database for manufacturing information. When this happens, a company might choose to store all of the information from the separate databases into a Data warehouse. Then the company uses a business intelligence program to get the information out of the data warehouse. Here’s a picture of a complex business intelligence architecture.

Remember this is all about how information flows throughout the company. The main difference between a complex architecture and a standard one is the number of operational systems being used. If there are a lot of operational systems, then a company will often create a data warehouse.

If you have a relatively small company and you can keep track of all your data in one system, then chances are your architecture will look like the standard architecture above. However, if you are part of a large organization and your company has to use multiple databases to keep track of all of the data that you have, then your business intelligence architecture will likely be very complex.

Most companies now see the value of mining through data to find useful statistics and trends. Company leaders use these trends to make profitable decisions. For example, a supermarket manager might notice that more people buy salsa when the salsa is placed right next to the tortilla chips. This makes sense, but the manager may not have thought of it until they saw the actual data. The act of mining through data to find information is known as business intelligence.

Keeping all of your data organized requires a lot more work than you might think. It requires a solid foundation and good structure. The way you organize your data is known as the business intelligence architecture of your company. Each company is organized differently, but here are four common components of most companies’ BI architecture.

(1) A sales system or database. When a customer buys something, he leaves behind important information. Information like customer name, address, phone number, and purchasing habits can be tracked in a sales operating system. Knowing information about your customers is often the key difference between success and failure.

(2) An accounting system. This is where all the important accounting information of a company is kept. Information on revenues and expenses are kept in this system.

(3) A manufacturing system. If your company actually creates physical products, then chances are you are going to have a database that keeps track of everything. There is a lot of information that travels through a manufacturing warehouse. You have your raw materials, your work-in-process materials, and your finished goods. Having a good system here will ensure that you don’t have shrinkage from employee theft or carelessness.

(4) The business intelligence front end. This is a key part of your business intelligence architecture. The front end usually refers to some form of software that you use to access the information. It is a computer program that you use to go through and organize all the data into understandable reports.

If any one of these four parts is missing or not functioning properly, then this can cause considerable damage to your business. For example, if your sales database isn’t functioning properly, this means you won’t be able to access information on customer purchasing habits. If you don’t know what your customers are buying, then you are going to have a hard time selling things to them. In order to get the most out of your data, you need a good, solid business intelligence architecture.