Once your business intelligence project has been scoped out, it is time to begin BI implementation. However, issues such as database compatibility and HIPAA requirements can make the implementation process a real challenge. This article discusses three issues that you might encounter as you begin a BI implementation project.
First, choose a BI implementation team
One of the first steps that you will have to take is to determine who will be on the business intelligence project team. Obviously, IT department employees will be the ones who perform most of the work, but your business intelligence implementation team will require members from outside of IT if it is to be successful.
There are two main reasons for this. First, you will need someone from each department who has access to the data that will be used by the business intelligence software. Sure, the IT department usually has administrative access to the servers on which the data is stored, but IT employees typically do not have access to the data itself.
The other reason you need representatives from multiple departments on your BI implementation team is because your business intelligence project may dramatically impact some departments (as will be explained later). If that happens, then your life will be much easier if you have someone on your team who works in the department that is being impacted. This person can go to bat for you when office politics rear their ugly head.
Finally, you will need to have a HIPAA expert on your implementation team. This person is responsible for making sure that your business intelligence project is not implemented in a way that violates any of the HIPAA requirements.
Second, master your master data set
One of the main reasons why BI implementation is such an ordeal -- particularly for organizations that don't have a data warehouse -- is because BI dashboards typically pull data from a variety of sources. A previous tip discussed the importance of establishing an authoritative master data set that can be used by a variety of applications. However, creating a master data set isn't always an easy process.
Typically a master data set includes data that is used by multiple applications and that is relatively static. Employee data from the HR database and patient data are a couple of examples of data that might be suitable for inclusion in a master data set. While this sounds simple, creating such a data set means overcoming several challenges.
The first challenge involves HIPAA compliance. HIPAA mandates that only certain types of employees have access to patient data. This shouldn't even be an issue when implementing business intelligence software, since BI looks at the big picture rather than focusing on granular information such as individual health records. Even so, if you plan to include any patient information in your master data set, you will have to ensure that the manner in which the data is stored, secured and accessed complies with HIPAA requirements.
BI implementation can have a major impact on the individual departments that own the data to be analyzed.
Another challenge is that not all data is suitable for inclusion in a master data set. Any type of data that is highly dynamic is generally considered to be an unsuitable master data candidate. In a health care organization, such data might include patient billing data or medical supply inventory information. Often, however, these are exactly the types of data that need to be analyzed by business intelligence software.
Business intelligence software does not require all data to reside within a master data set, but it must be able to access any data that is being analyzed. This requirement can pose a problem if the data presently resides in a proprietary database, rather than in a standardized database format such as a SQL database.
Finally, make sure hardware can handle BI implementation
Another issue that you will have to take into account is that, though business intelligence software usually presents information through a simple BI dashboard, the information being displayed is often the result of analyzing vast quantities of data. BI software can run some monstrous queries against backend databases. As such, you will have to make sure that your backend databases are running on servers with sufficient hardware resources to handle the additional workload caused by BI-related queries.
In conclusion, BI implementation can have a major impact on the individual departments that own the data to be analyzed. In some cases, it may be necessary for a department to begin using a different set of applications so that the underlying data can be stored in a database that is accessible to your business intelligence software. In other cases, the BI software's query process may cause other applications that depend on the data to have a slower response time.
In any case, the transition process can be painful for the individual departments -- and that is exactly why you need an advocate within each department. You need someone who can go to bat for you when the BI implementation process doesn't go as smoothly as planned.
Brien M. Posey, MCSE, is a Microsoft Most Valuable Professional for his work with Windows 2000 Server and IIS. He has served as CIO for a nationwide chain of hospitals and was once in charge of IT security for Fort Knox. Write to him at firstname.lastname@example.org.