Quality and Governance are Beyond the Data
Jim Harris in Books, Data Quality tagged Data Governance, Philosophy, Steve Sarsfield, Tony Fisher Tuesday, November 2, 2010 at 3:00AM Last week’s episode of DM Radio on Information Management, co-hosted as always by Eric Kavanagh and Jim Ericson, was a panel discussion about how and why data governance can improve the quality of an organization’s data, and the featured guests were Dan Soceanu of DataFlux, Jim Orr of Trillium Software, Steve Sarsfield of Talend, and Brian Parish of iData.
The relationship between data quality and data governance is a common question, and perhaps mostly because data governance is still an evolving discipline. However, another contributing factor is the prevalence of the word “data” in the names given to most industry disciplines and enterprise information initiatives.
“Data governance goes well beyond just the data,” explained Orr. “Administration, business process, and technology are also important aspects, and therefore the term data governance can be misleading.”
“So perhaps a best practice of data governance is not calling it data governance,” remarked Ericson.
From my perspective, data governance involves policies, people, business processes, data, and technology. However, all of those last four concepts (people, business process, data, and technology) are critical to every enterprise initiative.
So I agree with Orr because I think that the key concept differentiating data governance is its definition and enforcement of the policies that govern the complex ways that people, business processes, data, and technology interact.
As it relates to data quality, I believe that data governance provides the framework for evolving data quality from a project to an enterprise-wide program by facilitating the collaboration of business and technical stakeholders. Data governance aligns data usage with business processes through business relevant metrics, and enables people to be responsible for, among other things, data ownership and data quality.
“A basic form of data governance is tying the data quality metrics to their associated business processes and business impacts,” explained Sarsfield, the author of the great book The Data Governance Imperative, which explains that “the mantra of data governance is that technologists and business users must work together to define what good data is by constantly leveraging both business users, who know the value of the data, and technologists, who can apply what the business users know to the data.”
Data is used as the basis to make critical business decisions, and therefore “the key for data quality metrics is the confidence level that the organization has in the data,” explained Soceanu. Data-driven decisions are better than intuition-driven decisions, but lacking confidence about the quality of their data can lead organizations to rely more on intuition for their business decisions.
The Data Asset: How Smart Companies Govern Their Data for Business Success, written by Tony Fisher, the CEO of DataFlux, is another great book about data governance, which explains that “data quality is about more than just improving your data. Ultimately, the goal is improving your organization. Better data leads to better decisions, which leads to better business. Therefore, the very success of your organization is highly dependent on the quality of your data.”
Data is a strategic corporate asset and, by extension, data quality and data governance are both strategic corporate disciplines, because high quality data serves as a solid foundation for an organization’s success, empowering people, enabled by technology, to make better business decisions and optimize business performance.
Therefore, data quality and data governance both go well beyond just improving the quality of an organization’s data, because Quality and Governance are Beyond the Data.
Related Posts
Video: Declaration of Data Governance
Don’t Do Less Bad; Do Better Good
The Real Data Value is Business Insight
Is your data complete and accurate, but useless to your business?
The Diffusion of Data Governance
MacGyver: Data Governance and Duct Tape
Jack Bauer and Enforcing Data Governance Policies
Data Governance and Data Quality
Reader Comments (4)
Thanks for this post, Jim.
It is good to see that peoples' thinking is evolving.
Quality and Governance are definitely beyond data. In fact, it is probably true to say that the quality of the data is a barometer for overall quality and governance. If these are being achieved, the enterprises data will be good.
The next step is to get people to realize that Process does not create data! This is so important as many are striving to improve data quality by improving process - and failing!
It is Business Functions that create and transform data! So, it is in Functions that all data quality rules are applied and, hence, quality assured.
Also, because a Function can be a step in many Processes, if you assure quality at the Function level, it is assured in all Processes of which that Function is a part.
Regards,
John
In an old Robyn Hood movie, Herne the Hunter asks – ‘what binds the hunter to the hunted’. Robyn answers ‘the arrow’.
In this post, Jim rightly observes that four critical concepts (people, business process, data, and technology) are critical business concepts. But what binds people, process, data, and technology – what is the common thread to which they all belong? In my view, the anecdotal ‘arrow’ is business policy – the decision based framework that both causes and governs people’s behavior; that drives the need for process; that both consumes and creates data; and all of which is increasingly automated through technology.
Business policy can always be reduced to an underlying decision making strategy that determines how the business creates value in the marketplace. Ultimately a business’s decision making strategy (as articulated in its business policy statements) is the raison d’etre that binds the four critical concepts observed in the post.
The decisions that a business makes are its ultimate value statements - the people, process, data, and technology provide the means for both making and actioning those decisions.
To understand the decision making behavior of an organization is to understand why people do what they do, why processes exist, why data is important, and how and why technology is able to amplify value creation.
Decision making is the arrow which binds the four essential concepts observed by Jim.
From the LinkedIn Group for Data Quality Pro, Milan Kučera commented:
“Quality is a result of people work, their responsibility, improvement initiatives, etc. I think it is more about the company culture and its possible regulation by government. It is the most complicated to set-up a "new" (information quality) culture, because of its influence to every single employee. It is about well balanced information value chain and quality processes at every "gemba" where information is created.
Confidence in the information is necessary because we do many decisions based on it. Sometimes we do better or worse then before. We should store/use as much as accurate information. All stewardship or governance frameworks should help companies at change of its culture, define quality measures (the most important is accuracy), cost of poor quality system (allowing them monitoring impacts of poor quality information), and other necessary things. Only at this moment we would be able to trust to corporate information and made decisions.
A small remark to technology only. Data quality technology is a good tool helping you to analyze "technical" quality of data - patterns, business rules, frequencies, NULL or Not NULL values, etc. Many technology companies narrowed information quality into area of massive cleansing (scrap/rework) activities. They can correct some errors but everything in general lead into a higher validity and not information accuracy. If cleansing is implement as a regular part of ETL processes than company institutionalizes massive correction which is only cost adding activity and I am sure it is not a right place to change data contents - we increase data inconsistency within information systems. Every quality management system (for example TQM, TIQM, Six Sigma, Kaizen) focuses at improvement at the place where errors occur - gemba. All those systems require: leaders, measures, trained people, and simply - adequate culture.
Technology can be a good assistant (helper), but a bad master.”
From the LinkedIn Group for Data Quality Pro, Julian Schwarzenbach commented:
“Governance can provide a valuable mechanism to improve data quality approaches without necessarily trying to 'boil the ocean'. Whilst most organizations aspire to 'perfect' data, this is something that it is unlikely you will be able to afford/achieve.
An effective governance process provides the means to prioritize which areas should receive improvement effort (if at all) in line with strategic objectives. As new challenges and requirements arise, the governance function can prioritize these new needs alongside existing ones. Metrics are a valuable tool to support this, however, if the governance function get bogged down in the actual data itself, then governance will be far less effective.”