Mike FrenetteManager, IT PMO| Halifax Water (retired)Halifax, Nova Scotia, Canada
There has been a lot of discussion lately about data, how to make sure it is valid, what to do when it is not, cleansing, and so on.
But what about data governance to ensure a project or other organizational unit has permission to move data from its native source to a another place, such as a data warehouse, to allow for the use of data to allow for insights available from analysis and combination with other data?
And how do you ensure persistence of access permissions to avoid inappropriate exposure of data?
Has this governance structure (assuming you have one) created complexities for your project teams and organizational business units? Or has it created simplification? Saving Changes...
This is where the role of data owners is critical. Once an organization has inventoried its data assets, it can then define who has ownership over their handling and then those folks are responsible for ensuring that the quality of the data persists.
The challenge I've run into in a few organizations is insufficient education and support for data owners to help them fulfill this role.
Kiron
...
1 reply by Mike Frenette
Jun 15, 2024 3:20 PM
Mike Frenette
...
And of course, there is possible exposure of data once it leaves its source and is ensconced in a data warehouse or data mart somewhere. Redundant access permissions can cause many issues.
PMO Leader | Speaker & Mentor | Content Leader – PMOGA Latin America
Hub| Catholic University of UruguayMontevideo, Montevideo, Uruguay
Data must have an owner who is responsible for the validity of the data, where it is stored, who can access it, and who can modify it. Saving Changes...
George FreemanThought Leader | Author | Architect| Florida, United States
As usual, good topic, Mike!
Most organizations have no problem identifying the business owners for their respective data domains. However, the often amiss strategy element lies in interpreting said data, especially when taking full life-cycle perspectives for an entity across the domains. Stated differently, the perspectives of sales, marketing, service, finance, administration, parent company, and the like will often have different takes on the same question circulating the hallowed halls of executive management.
For example, a classical difference of perspective is one of customer profitability:
[A] When working with “finance types,” they prefer an all-in perspective that distributes gross margin across the customer and product base. Although everything foots and ties into their ledger, they just algorithmically distribute their fixed costs (bloated or not) across their customers, encumbering questions that relate to behavior and a customer’s direct contribution to profit.
[B] An “operational type” would want to understand the customer and product base from the perspective of contribution margin, which provides perspectives based on direct revenue and direct cost, exposing behaviors of a given customer and providing trends that feed strategic decision-making.
Both perspectives are valid in the example above but address different questions, the nuance of which often gets lost and debated.
----------------------------------
This example is one of hundreds that create “data quality concerns” within an enterprise. Although there may be actual data quality issues, perspectives, and interpretations often get cast as quality issues—what is an executive to do?
The bottom line is that bringing “interpretation governance” to an enterprise is a big lift compared to the ownership and access governance question, in my opinion.
George
...
1 reply by Mike Frenette
Jun 15, 2024 3:17 PM
Mike Frenette
...
Interpretation Governance. I love it!
As usual, George.... Excellent answer!
Saving Changes...
Mike FrenetteManager, IT PMO| Halifax Water (retired)Halifax, Nova Scotia, Canada
Jun 15, 2024 1:39 PM
Replying to George Freeman
...
As usual, good topic, Mike!
Most organizations have no problem identifying the business owners for their respective data domains. However, the often amiss strategy element lies in interpreting said data, especially when taking full life-cycle perspectives for an entity across the domains. Stated differently, the perspectives of sales, marketing, service, finance, administration, parent company, and the like will often have different takes on the same question circulating the hallowed halls of executive management.
For example, a classical difference of perspective is one of customer profitability:
[A] When working with “finance types,” they prefer an all-in perspective that distributes gross margin across the customer and product base. Although everything foots and ties into their ledger, they just algorithmically distribute their fixed costs (bloated or not) across their customers, encumbering questions that relate to behavior and a customer’s direct contribution to profit.
[B] An “operational type” would want to understand the customer and product base from the perspective of contribution margin, which provides perspectives based on direct revenue and direct cost, exposing behaviors of a given customer and providing trends that feed strategic decision-making.
Both perspectives are valid in the example above but address different questions, the nuance of which often gets lost and debated.
----------------------------------
This example is one of hundreds that create “data quality concerns” within an enterprise. Although there may be actual data quality issues, perspectives, and interpretations often get cast as quality issues—what is an executive to do?
The bottom line is that bringing “interpretation governance” to an enterprise is a big lift compared to the ownership and access governance question, in my opinion.
George
Interpretation Governance. I love it!
As usual, George.... Excellent answer! Saving Changes...
Mike FrenetteManager, IT PMO| Halifax Water (retired)Halifax, Nova Scotia, Canada
Jun 14, 2024 4:02 PM
Replying to Kiron Bondale
...
Mike -
This is where the role of data owners is critical. Once an organization has inventoried its data assets, it can then define who has ownership over their handling and then those folks are responsible for ensuring that the quality of the data persists.
The challenge I've run into in a few organizations is insufficient education and support for data owners to help them fulfill this role.
Kiron
And of course, there is possible exposure of data once it leaves its source and is ensconced in a data warehouse or data mart somewhere. Redundant access permissions can cause many issues. Saving Changes...
Hi Mike, yes, we follow a Data Governance strategy with clear ownership, role-based permissions, and documented approvals before moving data. We log all transfers and review access every 3 months, which Gartner says can cut compliance risks by up to 40%. Having standard rules for quality, retention, and security has sped up our projects by around 20%, as it removes approval confusion and rework. Saving Changes...
Community Champion
