Who’s the Data Boss? Anyone?

As a Data Mercenary, I’ve seen a lot. One thing that has regularly come up: disconnects in Data Governance which result in a lot of people feeling pain within and outside of the business, and confusion about who’s supposed to do what about what.

So, pay attention to this story because I don’t want this to be you.


We enter our tale and notice that the Department Head (DH) is frustrated, angry, and scared of losing a big account.

She approaches Really Hard Worker (RHW) and asks, “Why why WHY did this happen again? Big Client is calling about the same thing that happened last month. You didn’t send out the books to 3 people. Last month it was 5 people. WHAT IS WRONG?

RHW recoils in fear and stammers, “I-I-I don’t know. I receive the report. I follow the instructions that were given to me, and Big Client always calls about something. And they’re right. We owed them books for 5 people last month, and 2 people the month before.”

“You don’t know? Well, WHO DOES know?”

 ♠

The following month DH and RHW do the process together. They miss 3 people and Big Client is furious. “When are you people going to get it together?! I’m sick of dealing with this every month!”

The Data Boss!

Here’s the deal: the business rules changed

1998 Business Rule as Stated by Big Client

When employee completes courses 1 thru 4  they will receive courses 5 and 6 the following month.

2007 Business Rule as Stated by Big Client

Same as the original rule EXCEPT course 3 is no longer required in 12 states for employees who were hired prior to 2003.

The report isn’t capturing everyone.
It’s easy to conclude that the monthly report needs to be re-written.

The issue persists, however, because there is no DATA BOSS!

THE PLAYERS

The Consultant who writes the reports is expensive and works off-site. She receives instructions from the Department Head, writes and revises reports with Department Head approval. Big Client sets the rules and expects flawless execution.
Business Analyst runs the report and submits to Really Hard Worker
Department Head manages many business relationships and ensures that things get done. Doesn’t hear about the current problem until it’s persisted long enough for Big Client to lose faith in Really Hard Worker. Really Hard Worker has 2 days per month to follow a procedure, executing Big Client’s requirements. The rest of the month is spent on damage control, fixing problems, and looking for permanent solutions.
DID YOU NOTICE: No one is explicitly charged with regularly ensuring the integrity of the data? So, let’s ask:

  • When business rules are altered, added or dropped, who’s involved in the process of interpreting the rules?
  • Who’s familiar enough with the data to inquire about regularly occurring exceptions?
  • Who’s familiar enough with the business rules to inquire about regularly occurring exceptions?
  • What are some regularly occurring exceptions?
  • Who’s testing for strengths and weaknesses in the reporting?
  • Who’s involved with the testing of the reports?
  • Who has access to the source data that feeds the report?
  • Where do they fit in the hierarchy of the enterprise and how much impact do they have?

TIME FOR A DATA GOVERNANCE CONVERSATION
Formal discussions of Data Governance often sound like Data Governance is primarily owned by a company’s Business Unit and IT department. As they go on about the Database Administrator, Business Analyst, Data Analyst, etc, the problem is that these roles aren’t positioned where the rubber meets the road with the customers and clients. They find out about some issues after a lot of people have been quite upset for too long, flooding Customer Service with complaints.

Tom Jesionowski handles the topic better by being more inclusive. He describes Data Governance as being company-wide. It’s also a new and evolving discipline and we all are invited to contribute in defining the roles and responsibilities. So let’s think about it after taking a look at a couple of good articles.

A Simple Choice of Data Governance           Data Ownership vs. Data Stewardship


QUESTIONS

  • In the story about Big Client and Really Hard Worker, what suggestions would you have about the situation?
  • Is it bad to have an offsite consultant writing critical reports? She’s the only one who isn’t feeling regular direct pain.
  • What are the Department Head’s role and responsibilities?
  • Is there anything more that Really Hard Worker can do or could have done?
  • Should Big Client just chill out? Maybe Big Client should do the Data Management and just submit an order every month.
  • How would the Database Administrator or the Consultant get feedback before there’s widespread turmoil?

My Take

In the various definitions of Data Owner, Data Steward, Data Custodian, it’s not clear where a role like Really Hard Worker is defined. That’s the person who is most intimate with the data as it manifests itself in day-to-day business. RHW is the first person to hear the complaints. Also, RHW is often one who has little influence and limited technical and analytical experience. They might receive 3 days’ training and a 5-pg PDF of obsolete instructions.

We must look at the people we place in roles for execution of key processes and

  • Make Data Governance a clearly-stated expectation.
  • Designate them Data Boss: the point person for the integrity of the data.
  • They need to know that it’s not good enough to receive a report and assume that it’s fine or that the known problems are unfixable.
  • Provide training that’s consciously designed to overcome the company’s knowledge silos and develop sensitivity to areas where data are vulnerable for corruption and misuse.
  • The Data Boss has to be clear that if something is wrong, no one else is likely to find out until there’s a real problem. This requires having knowledge of the relevant players, their input and level of participation.

 

I’ve seen versions of this scenario many many times: a process managed by someone who has little influence, the stakeholders are unaware of the need for Data Governance, widespread assumption that the data and reports are impeccable … and then problems crop up and feel spooky.

Naaaah.

Declare, assign and empower a Data Boss. It’ll make a difference.