When some years ago we were offered to join to our first Business Intelligence project, we thought that something in the term was redundant because at the end of the day, doing Business requires Intelligence. This is the truth, especially if you pretend to do your business correctly, because profitable business cannot be performed without intelligence.
Speaking from a computer science book perspective Business Intelligence (BI) belongs to an Analytics World. Business intelligence is a set of tools and processes that helps you to make decisions based on accurate data, saving time and effort. The main idea behind a BI tool is the possibility of easily analyzable data based on business concepts without having technical knowledge about database tools or other sources that contain the data. BI tools pretend to extract knowledge from our stored data based in three main pillars: reliability, availability, and attractive user experience.
Imagine that you are the CEO of a small company dedicated to cookie manufacturing, and based on the sales reporting by product that you are analyzing, you detect that Cream Chocolate cookies have been decreasing in monthly sales every month. The figures you are seeing are now are about half of the amount that was selling at the beginning of the year. As CEO you have different possibilities: remove the Cream Chocolate cookie from the catalog; change the Cream Chocolate formula; set a bonus for your commercial department if they sell this product; or fire the brand manager because her department is causing you a loss, as reflected in the Profit and Loss analysis. But what happens if the real problem is that this product has changed its internal code because of a change in the cream formula and the product catalog is not correctly updated in your BI system—and it’s not reflecting properly the sales with the new code? Your previous decisions would be wrong because you are basing this on an incorrect data analysis. This is why reliability is a basic requirement in all IT projects but is especially relevant in BI tools because they can be used to take main decisions from strategic company management to basic operational activities. Based on that, it’s mandatory that data offered by our BI system must be consistent, and every analysis dimension must ensure correct results based on the data quality.
Now imagine that you are working on a car assembly line; it’s Friday at 8 p.m. and you need to make the order so as to refill your warehouse of different pieces before going home. You are launching your warehouse dashboard that suggests the quantity of every single piece that you need to ask about, and the source of this information has information from Wednesday afternoon because the refresh daily process has not finished yet. Next week you will suffer some stop on your assembly line because of missing wheels or you will have the warehouse completely swamped because you asked for 100 bumpers and they arrived last Thursday. A similar reason could cause the same result if you cannot access the system at the time required due to some maintenance activity on the platform, and you need to estimate the order based on what you think is missing.
Our system must be available to our users when our users need to use it. This seems to be an obvious condition, but there are two main factors that could cause us to fail to achieve this objective. Our system must be stable, running correctly during business hours, and data must be updated accordingly to our target consumers and their requirements.
A last main characteristic of the system that we expect to build is that our access to the available data must be user friendly and adapted to consumer expectation and capacity. You cannot provide expert analysis of data with a tool where they cannot interact with the information and, on the other hand, you could be in trouble if your commercial staff, which has no idea about computers, is required to publish a SQL query to get the data that requires analyzing. To be able to provide a tool that is user friendly, you need first to know your user and agree with their requirements based on their needs. Also a provided solution must be adapted to user capacity.
Achieving these three points is not an easy task, but they are the basis to delivering a profitable and lasting BI solution inside your company or for your customers.