By Elliot King
The data quality version would go–let me fix what I can fix; minimize what can’t
be fixed, and know the difference between the two. If you don’t understand what
data quality initiative can’t do, data quality stakeholders are bound to be
disappointed and those responsible for data quality initiatives will always be
on the defensive.
So what shouldn’t you expect from a data quality initiative. First, you should
not expect that you will be able to keep bad data out of your system. The
problem is that in many cases bad data does not “get in” your system from
somewhere else. As data ages, information that once was correct may no longer be
so. Moreover, data quality is about more than correct and incorrect data. It is
about being able to access correct information easily and in anticipated
formats. There will always be work to do.
Second, the goal of a data quality program is not just to insure the accuracy,
accessibility and appropriateness of your data. The ultimate goal of a data
quality initiative is to insure that the business processes, (including decision
making), that consume the data, run as efficiently as possible. Data quality is
a means to an end and not an end unto itself.
Finally, for a data quality initiative to work, companies have to discard the
idea that business units are responsible or that the IT shop is responsible for
data quality. Data is a shared corporate resource and therefore all the
stakeholders have responsibility.
Simply put–when it comes to data quality, nobody is perfect; we just need to fix
what we can and minimize what cannot be fixed. And very importantly, know how to
distinguish between the two. With all stakeholders contributing this is an