DQ-Tip: “...Go talk with the people using the data”
Jim Harris in Books, Data Quality, Events tagged Arkady Maydanchik, DQ-Tip, TDWI World Conference Thursday, October 15, 2009 at 1:38PM Data Quality (DQ) Tips is an OCDQ regular segment. Each DQ-Tip is a clear and concise data quality pearl of wisdom.
“In order for your data quality initiative to be successful, you must:
Walk away from the computer and go talk with the people using the data.”
This DQ-Tip came from the TDWI World Conference Chicago 2009 presentation Modern Data Quality Techniques in Action by Gian Di Loreto from Loreto Services and Technologies.
As I blogged about in Data Gazers (borrowing that excellent phrase from Arkady Maydanchik), within cubicles randomly dispersed throughout the sprawling office space of companies large and small, there exist countless unsung heroes of data quality initiatives. Although their job titles might be labeling them as a Business Analyst, Programmer Analyst, Account Specialist or Application Developer, their true vocation is a far more noble calling. They are Data Gazers.
A most bizarre phenomenon (that I have witnessed too many times) is that as a data quality initiative “progresses” it tends to get further and further away from the people who use the data on a daily basis.
Please follow the excellent advice of Gian and Arkady — go talk with your users.
Trust me — everyone on your data quality initiative will be very happy that you did.
Related Posts
DQ-Tip: “Data quality is primarily about context not accuracy...”
DQ-Tip: “Don't pass bad data on to the next person...”
Reader Comments (5)
From the LinkedIn Group for The Greater IBM Connection, Michael Goodson commented:
“I’m unsure why anyone would be perusing data without the specific objective of producing actionable information.
It would be foolhardy to not have a defined objective directly related to a process and an understanding of what constitutes a successful result before conducting any data analysis. If the variables required to answer a question are not defined, the actionable information will not be attained.
Data represent facts. The variables required to answer the question must be determined and the data must be located and manipulated in a way that provides the information necessary to make fact based decisions.
When the data are incorporated into the necessary logic to produce the answers being sought, information is the result.
Further complicating the issue is the quality of the data itself. If the gatekeeper responsible for the initial data input has not defined appropriate standards to assure the usefulness of the data for reporting purposes, the data will be flawed and the resulting actionable information may be incorrect.”
From the LinkedIn Group for DAMA International, Art Samion commented
"This is a pretty generic tip and one that I would hope most Data Warehousing professionals would regularly do in every Data Quality initiative.
I think that not only is it important to talk with the people that use the data, but also talk to two other groups of people:
(1) The people who create the data
(2) The sponsors of your particular project
From my perspective, understanding the data is essential. Understanding the process by which the data is created and how it fits into the big picture - is really what makes or breaks a good Data Warehousing individual."
From the LinkedIn Group for DAMA International, Marcus Jennings commented:
"Depends on how you define data user. In most cases, in my domain, I would say the end data user isn't the same person who entered or created the original data record and isn't, in general, the data owner.
In fact many end data users are unaware that their source data sets may contain errors. What can be gained from discussions with end data users is data value or business dependency.
Understanding data value can help in prioritizing areas of data quality initiatives."
From the LinkedIn Group for the IAIDQ, Tony O'Brien commented:
"I agree we have to 'involve' everyone - data producers, processors and consumers - if you've 50 sites or more, visit them all:
1. Explain the underlying reasons behind the data quality initiative
2. Identify where ownership (custody) and responsibility rests
3. Ensure your people are fully aware of the implications of their actions and how this effects their colleagues
4. Explain how this initiative will support everyone's overall objectives
5. Learn from the above and do it again"
From the LinkedIn Group for the IAIDQ, C. Lwanga Yonke commented:
"Talking with the people using the data is so crucial!
Here are a few more tips related to that first one:
a) Don't just talk with them. Also go observe how they are using the data. There are many things they do with it that they may never explain in words.
b) Also go visit those who create the data and those who enter it (if different). Observe their processes, note their constraints, obstacles, thank them, and see how you can help.
Also, if at all possible:
c) Take a group of people who use the data to go visit those who create and enter it.
d) And take a group of people who create and enter the data to go visit those who use it.
e) During these "gemba walks," engage the two groups in a conversation about data quality requirements and process changes (if any), so that the data "create and entry" processes can meet these requirements. Remove as many obstacles as possible.
e) Then celebrate!!!!
f) And repeat with the next subject area. You're just getting started :-)"