Data Management is More Important Now Than Ever

Robert (Bob) S. Seiner is the publisher of
The Data Administration Newsletter (TDAN.com), an internationally recognized and award winning online publication that focuses on the management of data, information, and knowledge as valued
corporate assets; he is also the Owner/Principal of KIK Consulting. Tony Shaw, Chairman of Wilshire Conferences, recently had the opportunity to ask Bob about his views on major trends and
challenges facing the data management community.

Wilshire: Bob, there’s a lot of talk on the discussion boards and in the trade journals about where data management is heading. You’ve been publishing TDAN.com for 5 years and
you have a pretty broad perspective on this subject. What’s your high-level outlook for where data management is headed as 2002 draws to a close.

Seiner: I believe data management is more important now than ever. To be honest with you Tony, while there should be greater demand for discipline
in data management, companies in general have not made spectacular improvements over the past five years. But I see that changing.

With the recent events on Wall Street – CEOs being forced to sign off on the accuracy of their reported numbers – I think you will see upper level management paying closer attention to the value of
data management. I expect many companies will begin to focus on auditable means of assuring managed data. The “information expectation” of these executives will drive them to become more
disciplined.

As the economy tightens, companies are demanding increased production from fewer employees. The result has been less time spent on improving how information assets are managed. Apparently many
companies feel they are “adequately” covered by their data management skills and they are “comfortable” with their level of data management discipline. However, many of these same companies
acknowledge that the cost of business data integration is a lot higher than it needs to be and the quality and consistency of their data assets is not where it needs to be.

So you have these converging trends, starting at the top of the organization with CEO expectations and concerns, and I think that means data management will receive increased corporate commitment
in future.

Wilshire: I know you had a strong reaction to the interview we recently completed with Graeme Simsion, in which he claimed that “Data Management isn’t Working”. Can you explain
why you feel he’s wrong?

Seiner: I believe that Data Management works. But companies cannot pay “lip service” to it and expect to reap any benefits. They need to have a
well thought-out plan, the time and resources, the tools, the executive support, and most importantly the patience, skills and knowledge to make it happen. These half-dozen items don’t always come
together but when they do, data management is working great at many companies.

Graeme made the statement that “traditional data administration seldom delivered on its promises in the past and is even less likely to do so in the future.” I would agree that poor data
administration or seat-of-pants data administration seldom delivers on its promise. However, I would hardly blame data administration or data management itself. Guess where that leaves the blame?

As for the future, … the future of data management is in the voice of the executive level of the company and in the hands of the knowledgeable and hard-working data practitioners. That
combination is required for success.

Wilshire: Well you guys will both be in Pittsburgh in early November at the Enterprise Data Forum, so maybe we can fix you up with some gloves and let you go at it. Meanwhile,
you’ll be talking about what it takes to go from “Good Data to Great Data” in Pittsburgh. What’s the inspiration behind this theme, and what major points will you be making in your talk?

Seiner: I don’t think you will see me squaring off with my friend Graeme any time soon. If he feels strongly that data management isn’t working
then I would rather he tell people that so they’ll wake up and say, “hey — you’re right, data management the way we have done it in the past is not working.” These same people should then set
out to improve how they manage all of their information assets.

The presentation of “GOOD (data) to GREAT (data): Why Some Companies Make the Leap and Some Don’t” is based on a 2001 best selling book of the same name (minus the “data” part) by Jim Collins.

In the presentation, I will start by talking about four traits of GREAT data (GREAT data must satisfy business needs, be viewed as an investment and not an expense; must focus on efficiency &
effectiveness; focus on quality) and how companies need to improve their planning and delivery in regards to five key areas to help them to move from GOOD (data) to GREAT (data). Those areas are:

  • Publishing and enforcing information policy
  • Applying accountability for information assets
  • Evolving the enterprise model as the basis for corporate data re-use
  • Concentrating on quality and measures
  • Managing meta-data in the lime-light and not in the back closet

The presentation includes case studies presented with two fellow Pittsburghers, Bill Lewis and Hal Davis. I am looking forward to sharing this presentation with the attendees at the event.

Wilshire: You just started your new consulting firm, KIK Consulting. Congratulations. And you’re near completion of your first major project, developing a data governance
strategic plan for a marquee client in the entertainment and hospitality industry. So what do you see as the key points to a successful data governance and stewardship program?

Seiner: Thanks Tony. KIK Consulting is focused on “Consultative Mentoring” – a cross between consulting and education. Companies have skilled
people working for them and often already own the tools and technologies that will enable them to be successful. Mentoring shows them how to succeed with what they already have, rather than doing
the work for them. As it is said, “Teach a person to fish and you feed them for a lifetime.”

There are three critical points that I stress when defining a successful data governance and stewardship program. I call it “Stewardship in 3-D”. The three Ds are “de facto”, “discipline”,
and “database”. I wrote an article about this subject in the present issue of The Data Administration Newsletter (TDAN.com).

The 3-Ds …

“de facto” – Stewards already exist in most organization and thus are “de facto” by nature. Individuals already have accountabilities for defining how data will be used-created-updated,
defining business and quality rules associated with the data, … yet these accountabilities are not clearly defined and the accountabilities are not formalized (or recorded) anywhere. Data
governance formalizes and records these accountabilities and makes use of this information as a value-add instrument for the entire organization.

“discipline” – Once stewards are accurately identified and recorded they must be used at specific “touch points” throughout the data and/or application development lifecycle. Stewards must be
actively engaged and held accountable for managing data in a manner consistent with an enforced and well-communicated “Information Policy”.

“database” – Steward information must be recorded somewhere. That somewhere is a stewardship or governance repository. Stewardship data is meta-data. The steward database houses data that relates
people in the organization to their steward roles and the data they steward. This database, a stand-alone repository or an extension of an existing meta-data repository, when kept up to date,
provides a roadmap to the data of the organization and becomes an important quality tool.

(See the full article at http://www.tdan.com/view-articles/5127 — Ed.).

Wilshire: Back to your days at Highmark Blue Cross Blue Shield, you were deeply involved in meta-data management. I continue to see lots of client interest in “meta-data”, but it
seems that the applications are increasingly diverse. What patterns are you seeing?

Seiner: Meta-data is not just “data about data” anymore. The definition (and my understanding) of meta-data has expanded greatly over the years.
Meta-data has become recognized as the backbone of all information assets.

The term “meta-data” is now used in association not only with data management, but with content management, knowledge management, document management, email management, business intelligence/data
warehousing, business rules, data governance and stewardship, … wherever there is a need to use information you can find meta-data.

The term “meta-data” is now associated with every tool — from data modeling tools, to data integration tools, to data query tools, to data quality tools, to knowledge management tools, to data
mining tools, to document management tools, … and the list goes on, every tool that houses and manipulates data uses meta-data.

The pattern I am seeing is that companies are still struggling with how to put this meta-data to use. To limit this struggle companies can focus on two areas … 1. Do a quality job of capturing
the meta-data in the tools (i.e. avoid entering cheese-burger definitions – a burger with cheese) and 2. Find ways to get the meta-data out of the tools and into the hands of the business and
technical people that can use this information.

Wilshire: So, would you agree that meta data is becoming the “glue” that joins the worlds of structured and unstructured data? Where is this convergence leading us next?

Seiner: Meta-data is the glue and we are the carpenters. I have no problem stating that meta-data is the backbone of tying together all
information assets.

Companies are already looking to merge the worlds of structured and unstructured data. Look at effective e-business applications, customer relationship management, enterprise portals, wireless
technologies, … they are already heading this direction. Companies are quickly finding that the most efficient and effective ways of merging structured and unstructured data involves the use of
meta-data or what they know about their information assets.

Whether the discipline is modeling, governance & stewardship, integration & data movement, quality management, content management, knowledge management, the list goes on …, meta-data
exists at the core of the discipline.

I would only add one thing … I think that meta-data and governance (applied accountability) together are the two “glues” that hold structured and unstructured data houses together.

Wilshire: You’ve anticipated my next question, which is…what is the role for the classic DA function (i.e. data administration, data architecture) in current business
initiatives such as Knowledge management, Business Intelligence, CRM, ERP, etc.?

Seiner: The “classic” DA should play a role in all of the initiatives you mentioned. If there is a data component to an initiative, either
structured or unstructured, somebody who has the enterprise’s data interests in mind should be involved, someone should be accountable for defining specific requirements for how the data will be
used, and someone should build a knowledge base surrounding the information about what makes up that data component. OK … so did I digress back into stewardship and meta-data? Is there a trend
here?

The initiatives you mentioned, at least BI, CRM & ERP, all focus primarily (although not solely) on structured data. Knowledge management focuses on both structured and unstructured data. I
have seen companies involve the DA role and not involve the DA role in these types of efforts. It is a “pay me now or pay me later” proposition. Paying later always costs more.

Share

submit to reddit
Top