Why You Need a Data Audit
In many cases, the impact of bad data is not too hard to identify–direct marketing campaigns with high numbers of email bounce-backs or undeliverable mail; marketing efforts with poor response rates because the offering has not been correctly tailored to address the “wants and needs” of the customers; prospects who turn away because their names are spelled wrong or other information is incorrect.
With the growing need to integrate data from so many sources, both internal and external as well as the dynamic nature of business itself, the chances of introducing mistakes into critical databases is growing exponentially. It’s not really a question if there are mistakes in your databases; the question is how many mistakes are there and what kind of negative impact will they have on your business?
The answer to those questions can be determined by conducting a data audit. A data audit is just what the name implies–a systematic look at data to insure that it is what it is supposed to be.
Conducting a data audit is pretty straightforward. First, determine which records and which fields should be examined. If customer records are to be audited, fields such as names and addresses perhaps; buying history, payment history or other critical fields may be included. The field selected should be those associated with a specific process or activity.
Then assess the fields to determine which do not conform to the data dictionary and the business rules that govern them. A wide range of errors may surface. Values may be missing. Values (such as dates) may fall outside of a specific range. Fields may be incomplete or formatted incorrectly.
The next step is to determine the source of the error; how the errors can be rectified (if, in fact, they need to be rectified); and how mistakes can be minimized in the future. Among the most common sources of errors are faulty data conversions and inconsistencies in integrating data from different sources. Not every error always has to be fixed or the underlying process changed. In most situations, there will always be some errors in the data that can be tolerated because they do not have a crippling impact on the business activity being supported.
How often should data audits be conducted? Not surprisingly, the answer is elastic. Data audits should be conducted often enough that faulty data does not impede reaching your business goals.