Stop me if you’ve heard this one before… Garbage in equals garbage out. Somehow, it’s a phrase that never goes out of style.

As you design and implement BI projects, it pretty much goes without saying that results will be only as good as the quality of the data fed into the system at the start of the project. The value of your business intelligence is contained within source data.

Now consider what happens as you enrich or append your existing information with data from other sources. Do they use the same business standards and rules that you’ve set up for your data? Your value chain extends out to that appended data as well. For the same level of value at the end of the process, BI system designers need to pay as much attention to the art of linking their data to third-party sources as they do to their original records.

Consider how and why data enhancements are applied to business intelligence databases.

First, most customer data repositories typically contain names, addresses, phone numbers, email addresses and product(s) or service(s) purchased. Maintaining the accuracy and timeliness of those fields – data quality – is vital. But often managers want to know more than the data before them can tell, so they pursue other demographic details or corporate parameters that can be appended to original source data.

For data enhancements like this to be fully effective, data must be standardized for effective linking of demographic information. If ‘Bob Johnson’ is in the original database, it may never link to ‘Robert Johnson’ in the demographic database without effective data quality tools. Users must set up the business rules and decide which fields are used to link records and make decisions about how loose or tight the relationships are.

Secondly, corporations need to manage sales territories in order to effectively and fairly parcel out which sales leads go to which sales person. To assist in this process, incoming leads can be enriched with additional data appends, including longitude and latitude. Demographic appends can also offer further valuable information, such as whether the lead is on a “Do Not Call” list, size of household or business, SIC codes, public or private company, household income, etc. In a perfect world, the lead can be simply distributed based on territory. However, in most companies, corporate executives reserve some of the super accounts for themselves and not the territories. Therefore, a lead coming in first must match against the corporate super account list (another level of data enrichment), then into the territories. In this process, systems also must recognize IBM, I.B.M., International Business Machines, and Lotus (an IBM company) as all one company.