A savvy IT professional will ask end users about their requirements, before trying to introduce business intelligence to his or her organization. They will ask questions like: "What information are you looking for?" and "How do you want it to be formatted?".
1. What level of technical skill do the users already have?
This question should not be posed directly to users. Many end users have a somewhat optimistic perception of their technical abilities and most consider themselves more of an expert than they actually are. As a general rule of thumb, if you are handy with Excel, i.e. can use formulas and sort data, and you can interpret numbers with relative ease, you are likely to be able to comfortably work with a business intelligence solution.
2. How much time can you dedicate to finding, accessing and analyzing information?
This is a pretty essential question, probably the most important of the five, but is often completely ignored. Yet if it is answered realistically, you can easily create an accurate deployment plan, even without responses to the other four questions. In most companies, people need access to information, but they often lack the time to dedicate to it. If you don’t have a lot of time to spend finding, accessing, and analyzing information, then you probably don’t require business intelligence capabilities. How much time someone has to devote to reporting is often related to their role in the organization. For example, CEOs may have very little time to spend analyzing data, while analysts will have more since analysis is what they are primarily responsible for.
3. What types of questions will users be asking?
Different types of questions are best asked and answered by different types of solutions. Performance management, dashboards, and scorecards are about monitoring the status of a key metric, whereas ad hoc query tools are better suited to asking random or on-the-fly questions.
4. How timely must the data you are accessing be?
Your BI deployment will be influenced by the required latency of the data involved. Business intelligence has three possible levels of latency – scheduled update, on demand, and real time.
With scheduled update, the information is not up-to-the-minute. The data source is probably updated on a scheduled basis, e.g. once a day or every hour. Or, the data can be close to real time, and be updated incrementally within minutes of a transaction. The majority of BI deployments work well with some kind of a scheduled update.
Users will require direct access to information contained in a near real-time data warehouse with an on demand solution. While this does not necessarily affect the BI deployment itself, it is uncommon for BI tools to give users access to this kind of information, because it is usually stored in a form that makes creating ad hoc queries difficult. However, dashboards, reports, and guided ad hoc reports are effective business intelligence deployment methods when on demand access is required.
Finally, true real time means that the information being displayed is updated as an event happens – even before the relevant data makes its way into a database. Only dashboards and reports will enable the dynamic display and update of information in real time.
5. What role does the user play within the organization?
Understanding a user’s role gives you a certain insight when it comes to their BI requirements. For example, executives look primarily for key performance indicators, whereas analysts will often spend their time analyzing data in more detail. A front office worker, such as a customer rep, often searches for information to solve specific problems. While there are likely to be differences from one organization to the next, understanding the characteristics of the various users in your particular firm will help you better understand user requirements.