Contact Us
Call us
United States
+1 646 893 3042
Accounts receivable department
+1 646 974 0772
All other
+32 2 793 02 19
North America: USA and Canada
Collibra Inc.
61 Broadway, 31st Floor
New York, NY 10006 - USA
EMEA: Belgium
Collibra NV
Picardstraat 11 B 205,
1000 Brussels - BELGIUM
View all
Register for access
Register for access
  • Dashboard
  • University
  • Data Citizens
  • Marketplace
  • Product Resources
  • Support
  • Developer Portal
By signing up you agree to Collibra's Privacy Policy.
My Profile
John Smith
Data Scientist, USA
Cloud-Ready Data
Digital Transformation
Data Governance

Still can’t find or trust your data? Tell Bob goodbye.

Can't trust your data?

The beginning of the new year is always a time of reckoning. A time when organizations ask themselves what worked, what didn’t, and, most important, how to use that information to pursue new strategies and ventures.

For some organizations, often those without a clear understanding of the metrics driving their performance, that day of reckoning can be brutal. A recent assessment by The Economist of General Electric’s publicly available financial data is a case in point. With no consistent measure of performance, it has become difficult (I believe the author would argue impossible) for the company to answer simple questions about its assets and liabilities. Consequently, its new CEO faces tremendous hurdles as he tries to turn the company around.

In 2018, data will continue to be a differentiator. But many of the organizations I speak with are still struggling to master it. These organizations understand the importance of data. And when it comes to their own data, their priorities are sensible—even laudable. Most cite improvements in data confidence, data protection, data management, data recency, and connected data sources as their priorities for 2018.

But their data scientists and analysts—the folks who should be spending their time answering pressing business questions—still spend as much as 80% of their time searching for trustworthy data.

If organizations understand that data is the new coin of the realm, why is it so hard for them to use it well?

I think I have the answer. It’s all about Bob.

Who’s Bob, you ask? Bob is the fellow who knows his data. He knows how to source and combine data to answer your questions. He understands where it comes from and what’s been done to it. He’s scrubbed the data himself and understands the rules he applied to it. If you happen to know Bob, you know he’s a good source for the information you need.

Organizations love their Bobs. And with good reason. Bob is the embodiment of tribal knowledge. Sometimes it seems like he can work magic and pull the answers you need out of thin air.

The problem with Bob, of course, is that his concerns may not always be your organization’s concerns. Bob may be a fanatic when it comes to data quality, but have very little understanding of data protection. I would lay odds that Bob really doesn’t care too much about lowering the costs of data management across the organization. He may not have access to most recent data or the right data sources. And even if you have confidence in the data Bob is providing you, that confidence is based on your feelings about Bob, not on any transparent and shared measure of data quality.

This is how tribal knowledge works. It’s a system of unwritten (and so unverifiable) rules that can shift and change over time. I know that tribal knowledge can move projects forward—I’ve relied on it myself on more than one occasion. But for organizations to rely on their Bobs as their system of record is dangerous. Take a deeper dive into that Economist article I mentioned and you’ll find out that GE uses 18 different definitions of cash flow. I don’t claim to have a definitive understanding of what’s happening at GE, but I’ve certainly seen conflicting definitions of important metrics derail business strategies. And that’s what happens when an organization depends on its Bobs rather than on a shared system of record.

Organizations who have a data governance platform with a governed data catalog in place no longer need to rely on their Bobs for the information they need to make critical decisions. Instead they can empower Bob, along with Sarah and Rahel and Jian and their teams of analysts and data scientists to more easily find the data your organization will need to excel. With a shared data governance platform, everyone across your organization will know where to find the right data, they will have a library of metrics and data that provide a common understanding of what that data means, and they will be able to trust that the data they’re using is accurate and fit-for-purpose. Collaborative data governance is one very good way to systemize tribal knowledge. I think even Bob would approve.



Related resources

View all resources

More stories like this one

Nov 10, 2021 - 2 min read

Ensure faster, smarter and more trusted analytics with Collibra and Tableau

Read more
Jan 11, 2021 - 3 min read

Enabling trusted Business Intelligence with Looker integration

Read more
Jan 6, 2021 - 3 min read

What is an augmented data catalog?

Read more