As anyone who has seen me present or any regular reader of this column knows, I'm fond of using analogies, quotes and stories to make points come to life. A colleague who wondered how I came up with all of these gems asked me about this once. My answer was that data warehouse stories are all around us if you just open yourself to the experience.

And so it is. With nearly 20 years working in this industry and helping others achieve the success and value of true data warehousing, I find that I'm often listening to shows, reading books or hearing the Sunday sermon and finding parallels that can be used as lessons to teach best practices (this is a blessing and a curse, as USA Network's Adrian Monk would say). For example …

"I don't care what you believe, just believe in it." —Shepherd Book, Serenity, 2005

This quote from Joss Whedon's space cowboy flick, Serenity, gets to the essence of leadership and governance—you must believe in your goals. I speak with conviction because I've seen that it works and I know what it takes to make a data warehouse successful. Based on this experience, I'm able to give advice and guidance from the position of a true believer.

This is important because the basis and foundation of the data warehouse governance is leadership, which means that you have to believe in what you're advocating. If you don't believe in what you're saying, then why should anyone else?

Many executives will espouse the idea behind bringing data together and having a cross-functional approach to data management. Often, however, those same executives will then approve projects and processes that are diametrically opposed to success. They'll also allow political barriers and turf wars to rage on when it's apparent that those battles are root causes for delay or inactivity in evolving toward an enterprise implementation.

It takes more than just saying you support the data warehouse—you have to mean it. The data warehouse governance team must believe and act in a way that focuses on the data warehouse and enables the company to move forward.

"It's supposed to be hard. If it wasn't hard, everyone would do it. The hard ... is what makes it great." —Jimmy Dugan, A League of Their Own, 1992

Too often, people want to have all the benefits without putting in any effort. In data warehousing, the easy part is getting the data loaded. The hard part is attacking all the politics about data sharing, process re-engineering and the like that constantly raise barriers and prevent a company from moving beyond the status quo.

I'm reminded of a relatively new customer speaking at our PARTNERS conference in 2005. Someone asked how the customer accomplished the difficult task of migrating historical data into the new model. The answer was straightforward. The presenter said, "We knew it had to be done so we did it, even though it was hard." Now after doing the hard data integration and migration work, the company has very quickly taken its data warehouse further than other companies and in about half the time.

Yes, many parts of implementing a true data warehouse are hard, so what is your point? Many things in life are hard and yet we make the decision every day to overcome the difficulties and proceed. If we did only what was easy, then we'd never reap the benefits or the joy of success.
 
So what are the "hard" problems you face in the data warehouse? Do you understand all the options and their consequences? Can you attack this problem in pieces or does it need to be a single hit? What are the business requirements and the value of meeting those requirements? Is the governance team actively working to address the non-technical obstacles? Are the right resources and incentives being brought to bear on the task? We have to deal with hard issues every day, but what doesn't kill us only makes us stronger.

"Failure is not an option." —Gene Kranz, Apollo 13, 1995

Before starting any data warehouse endeavor, you must decide whether you're willing to do what it takes and be committed to the goal. This isn't a one-time event; this is a foundational change in how your company treats and utilizes data. How are you going to react when the going gets tough? What are your barriers and how do you work together as a team to overcome them?

In the movie Apollo 13, the astronauts faced unimaginable difficulties. Everyone on the ground was exhausted, but they soldiered on because it was up to them to get the Apollo 13 crew home. The ground crew understood that they had to do everything possible to make sure the astronauts made it back safely.

A retailer I was working with had a similar—albeit less life-threatening—crisis early on in the implementation of its "perpetual inventory" application. This was something that had never been done before to this degree. As one would expect, there were challenges and unforeseen problems. Many on the project team were bemoaning the delays and additional costs. It got to the point where a meeting was called to get the project back on track.

At that meeting, one of the retailer's executives said, "This capability is worth three million dollars a day to us. It cannot fail. Do whatever we need to do to make this work." After that little "pep talk," renewed focus and perspective were brought to bear on the challenges. By understanding the value of our efforts, we saw that failing to overcome the data warehouse challenges would be fatal to the business.

Does everyone on the data warehouse governance and implementation teams understand—as did the Apollo 13 ground crew—that failure isn't an option? As Yoda would say, "There is no try; there is do or do not."

"Now, go do the right thing." —Dr. Laura Schlessinger, the Dr. Laura radio show

I am often amazed at some of the architectures and solutions I come across. In one situation, I saw that the extract, transform and load (ETL) layer was composed of several dozen SMP servers, and data was being passed around the architecture and replicated at least five times. During a review session, every point that was brought up about issues like consistency, replication cost and security holes was met by a reasoned response on why the process could not be done any differently.

Exasperated, I finally said, "Surely, you cannot think that this is the right way to achieve your business objectives?" Their response said it all. It was, "You're right, but unless we reuse the existing servers, we won't get the funding to go forward."

Now I don't blame the architects for this situation. They were all smart people working hard under tight time and cost pressures. They were told to build something that worked using the tools they had available, so they did. But in the long run, although they'd technically be successful in meeting their goals, they would fail to meet the real objectives of a data warehouse—an environment that is resilient to and encourages business change.

I told the architects that I didn't blame them for the architecture, but that I did hold them accountable for failing to confront the tough issues and tackle the barriers that prevented them from doing the right thing. IT may not be the right place to solve the problems, but they could at least explain the consequences of doing things the wrong way to those who can solve them.

I've often said that implementing a good data warehouse isn't difficult, but it is time-consuming. It's about doing what you already know is the right thing to do. It's about resisting the urge to do the easy over the necessary.

Of course, you must attend to the reality of the challenges facing you, but you need to attend to them with consistency, cooperation, communication and a drive to get this right for the company, not just for yourself. That is, after all, what governance is all about.

To close, remember what Uncle Ben told Peter Parker in Spiderman:

"Remember, with great power, comes great responsibility."

Now go use that power wisely. T

Rob Armstrong is a director with Teradata. Since 1987, Rob has consulted with or led teams to deliver some of the most successful data warehouse systems in existence.

� Teradata Magazine-September 2006