As regular readers of this blog know, I sometimes share answers to questions I get via e-mail (and other avenues). By doing so I can hopefully expand on the audience participating in the discussion…
Anyway, I received the following question via e-mail:
When trying to get management on board with a data quality initiative, what should I focus on in order to justify the investment? How do I build “the business case?”
This was my response:
The best (short) answer that I can to this question is to try to quantify the cost of poor quality data on the business. This needs to be written in business language and not technology-speak. For example, what is the cost of a lost sale because product information was incorrect? Do you have a way to identify these cases? Even anecdotal evidence can be powerful if you are talking to the manager of the product line that lost the sale.
A more in-depth example might be an approved (or unapproved) business decision made based upon the wrong data and assumption. What is the cost, in terms of lost business? Or a hiring decision made based on erroneous information. What is the cost, in terms of an ineffective work team (plus the cost of removing and replacing the bad hire).
The cost of poor data quality is pervasive… and it can be difficult to quantify effectively.
I realize that the cost of finding these problems can be enormous, too. It can help to have some industry expert help. I would recommend that you purchase and read any of the several excellent books that Thomas C. Redman has written. These books focus on the data quality problems and have some facts and figures on average cost of poor quality data to business. For more in-depth and technical treatments of data quality issues I would direct you to books written by Jack Olson and Larry English.
…I realize that this is a quick and dirty answer to a complex question, but that is usually all I can afford to do with e-mail Q+As. Did I get the basics covered? What would you have answered?
Please share your thoughts and comments…