By Elliot King
The first mistake companies make as they attempt to manage their data is that
they don’t take the time to understand exactly what they want regarding their
data quality. Completely mistake-free data is an unreasonable goal and probably
unattainable in most cases. Since data is constantly flowing into organizations
and being manipulated within business processes, errors will occur.
How tolerant of dirty data can your organization be? Which parts of the
organization can best cope with errors? In which parts of your processes is it
absolutely essential to have very accurate data?
The goal of a data quality program should be to minimize the negative impact of
faulty data. Dirty data should not inflict an intolerable cost on the
organization or impede critical business processes. The clear identification of
this goal is essential because, if you don’t know where you are going, you will
never know if you get there.
The second mistake is failing to assemble the right team. Data quality is not
solely a technical issue. It is a business problem and a corporate challenge as
well. Data quality programs rely on buy-in from all the stakeholders and often
needs sponsorship from the highest levels of the organization to succeed.
Next, companies should not try to “boil the ocean.” In most cases, you cannot
fix everything at once. When it comes to data quality, you cannot even identify
all the problems at once. Prioritize and triage. Identify and correct the
critical data issues first. And remember, determine the source of your faulty
data and correct that initially; before you attempt to remediate the symptoms of
your broken data entry processes.
The most enduring mistake a company can make as it tries to implement a data
quality program is to fail to recognize that data quality management is an
ongoing process. There are no quick or one-time fixes. Not only is it an ongoing
process, it touches many different parts of the organization and may require
people to change their work routines. And of course, the biggest data quality
mistake a company can make is to not have a data quality management program at