As the cab whisked me past New York, Rio, Paris, and assorted pyramids, I knew that I wasn’t in Kansas anymore. In the midst of the nightly sinking of the Titanic, rhinestone studded showgirls,
and cheap slots – over 700 serious, dedicated professionals came together in the middle of the desert to flesh out their knowledge of data warehousing. No, this wasn’t Burning Man, this was The
Data Warehousing Institute’s Winter Conference in Las Vegas, and it was the biggest TDWI event since the IT crash.
TDWI tried something a little different at this conference. At the unbelievable hour of 8:00 on a Sunday morning, Dave Wells kicked off the conference with a primer on a framework for data
warehousing and business intelligence. In spite of the hour, Wells had a good draw, and he made it worth our time. In 45 minutes, Wells led us through a process of constructing a framework that
placed all of the different activities associated with BI and data warehousing in the context of service to the business. In the end, attendees had a way to relate all of the different courses that
the conference offered throughout the week and more importantly, a way to relate all of the moving parts of this discipline for the folks back home.
I knew that I would be dragging Sunday morning so I put Michael Gonzales’s course on Business Analytics on my dance ticket. Gonzales brings a lot of energy to his classes and this was no
different. He quickly set about defining the core components of a successful BI delivery strategy. He talked about the place of portals and meta data and then led us through a breakdown of
functions that included static reporting, interactive reporting, OLAP, ad hoc query, data mining, and finished with a great section on spatial analysis. He also gave me a great tip on a good jazz
It was early in the week, but what the heck, good jazz is fine anytime. It turned out that Napoleon’s was in the Paris, which was right next door (only in Las Vegas). They served champagne and a
tasty quartet led by a great sax player, Tommy Thompson. The group was tight and I closed up the place listening to them. What a find. Check out http://www.tommythompsonproject.com.
In spite of starting out the week with not much sleep, I was anxious to catch Jill Dyche’s keynote, ‘From Platform to Portfolio: Evolving Your BI Program’. I had attended one of Dyche’s classes
at an earlier TDWI conference and was impressed by her insights into what managers need to know about data warehousing. Her book on customer relationship management is one of the best reads on the
subject. This morning she gave us pointers on pitching BI to the execs. Dyche’ stressed the importance of making that pitch in business terms.
Later on, I headed Howard Spielman’s ‘The BI Endgame: How to Win-Both Strategically and Tactically.’ Spielman used the analogy of an endgame in chess to get across the point that after all of
the work has gone into acquiring, transforming, and otherwise preparing our data, the value of what we do all boils down to the perceptual interface between the data as it is displayed and the
human brain. He cautioned on how graphics can be used to persuade rather than present reality. To underscore the point he broke the class into teams. Each team was assigned the task of preparing
graphics that would support one of the following: the sales manager, the CEO, the anti-trust prosecutor, the defense attorney, and the government watchdog advocate. He then gave each team the same
sets of data. Without much effort each team came up with a persuasive graphic representing its point-of-view. Spielman is brilliant and his classes set the bar for all others. Catch him at a TDWI
conference if you can.
That night I dropped in on Dan Linstedt’s class on Data Vault Modeling. Lindstedt has developed an approach to modeling the central data warehouse that is pretty revolutionary. His approach
diverges from standard 3rd normal form and dimensional modeling practices, but he claims that it is much more effective at the distribution level in reacting to requests for new data marts.
Lindstedt has written several articles on the Data Vault for TDAN. Check out his latest under how-to articles.
After night school, Cognos’s hospitality suite offered a welcome stop for refreshment and a chat with one of their marketing folks. Cognos has been actively promoting their enterprise planning and
enterprise scorecarding offerings. They also plan to be pretty active at the next TDWI conference in Boston, which will feature Robert Kaplan one of the creators of the balanced scorecard.
I always like to go to TDWI’s Business Intelligence Strategies sessions at every conference because they tend to focus on the current hot topics. TDWI titled this one: ‘Right-time Business
Intelligence: How to Create an Event-Driven Organization.’ The session consisted of several analyst and several case study presentations. Colin White went over his recent TDWI study, ‘Building
the Real-Time Enterprise.’ He pointed out that 13% of those firms surveyed had deployed real-time features in their data warehouses and 37% planned to do so. Looks like this is a growing thing.
Werner Enterprises, Continental Airlines, BT (British telecom company), and Brocade all presented RTDW cases. However, I ducked out mid-morning to catch a guru audience with the father of data
warehousing, Bill Inmon.
Inmon had slipped into town not to give a keynote or class, but to just spend the day giving one-on-one one hour guru sessions. I signed up to ask him some questions about the history of the data
warehouse concept. He recounted how he and a couple of other consultants were sitting around after work one evening in 1984 in Edmonton. They were working with separate clients in different
industries, but they had come up against similar decision support issues that centered on a lack of integration. They kicked around what to call the concept of an integrated decision support
database and came up with several ideas, including ‘atomic database’, ‘information warehouse’ and ‘data warehouse’. Inmon published an article on the information warehouse not long after
that, but then later on IBM started using the name to describe a technical architecture. Inmon and IBM had some discussions that Inmon can’t talk about, but he did go out buy a Porsche afterwards.
After that Inmon did a trademark search on ‘data warehouse’, which came up negative and then started using the term in his books and articles.
The vendor area opened up on Tuesday and extended into the evening so I dropped in to see what was going on. I saw a good iWay demo for their adapter software. Adapters should be a hot area with
the increasing interest in real-time data warehousing as organizations replace their batch processes with low latency feeds from disparate data sources.
TDWI has started to sponsor specific industry tracks at its conferences, and this time it was heath care. Wednesday morning I attended a half day session given by Robin Smith, Susan Easton, and
John Haughom from PeaceHealth titled ‘Patient Care Quality through Data Warehousing: A Case Study.’ Haughom was an MD who had given up his practice to become Senior Vice President charged with
improving patient care quality throughout the PeaceHealth system. I have never heard a more compelling story about the value of data warehousing than these presenters made that morning. Before the
advent of an effective data warehousing system, Haughom said that senior management tended to focus on the financials of the operation. Once they began to receive patient data from the data
warehouse, their focus turned to improving the health care delivery system. Where else could you compute data warehousing return-on-investment in lives?
The night ahead looked like it was going to be pretty long so I rented a car for the afternoon and drove out to Red Rock Canyon. Although Las Vegas has become a giant amusement park for adults,
there are some natural wonders to be had just outside of town. Red Rock Canyon is a national recreational area with a number of short hikes and a 13 mile loop road through some great scenery. I
hiked out on one of the trails about a mile or so and played my Native American flute for an hour. Now I was ready to go back to the bright lights and TDWI night school.
Back at the ranch, Jennifer Hay gave her class something to think about with her presentation on ‘Business Services and Customer Care for Business Intelligence.’ All too often data warehousing
efforts become data centric and overlook their reasons for being. Hay did a nice job paralleling IT and customer views of the development life cycle and then gave some suggestions for making the IT
organization more responsive to customer needs. Apparently, she struck a nerve. At the end of the class, one of my classmates hurried towards the back of the room to pick up extra hand outs. I
heard him count, one for my boss, one for my boss’s boss, one for her boss, etc. etc. etc.
Right after night school let out. Teradata sponsored a hands-on lab that I was keen to sit in on. I had some experience as a Teradata DBA over 10 years ago. Now, I wanted to bring myself up to date
since Teradata has been gaining a lot of attention in the real-time data warehousing arena. Sometimes labs of this sort are disasters, but Teradata did this one right. They started off the session
with a great dinner, and after a brief overview, they handed us a series of short demo scripts to follow according to our individual interests. They also had a number of high powered techies to
help us out with personal attention. By the end of the evening, I felt that I could step back into a Teradata DBA job without much hesitation.
Larry English started off the day with the second keynote of the week, ‘Can You Have BI Without IQ?’ Saying that we need information quality is like wrapping yourself in the flag. However,
English showed us how easy it is to skip over quality issues that can threaten the understandability of our BI delivery systems. It is one thing to have sound data quality reflected the data values
in our data warehouses, but English went further and showed how information quality depended upon the things that we often think of as meta data including names and definitions. He did talk about
data quality and the importance of driving data quality issues back to the source systems. However, English said that data cleansing alone cannot lead to a culture of information quality. He made a
number of suggestions on how to build up this culture including creating and maintaining a data mart of customer and knowledge worker complaints plus project feedback. You have one of these, right?
After the keynote I went off to the ICCP Exam Cram. TDWI is now providing time at its conferences to become Certified Business Intelligence Professionals through the ICCP (Institute for
Certification of Computing Professionals). Friday was exam day and Thursday, Pat Cupoli from ICCP gave us background information about the tests and gave us some sample tests. The CBIP criteria
include four years of professional experience (up to two years can be waived with a Bachelor’s degree) and passing three exams with a score of 50% or higher for the practitioners’ level and 70%
for the masters’ level. These exams consist of a core ICCP exam, a data warehousing exam, and one exam from either IT Management, Business Information Systems, Data Resource Management, Systems
Development, or Systems Security. Over the course of the day we got to take sample tests in each area. These exams were not easy and we knew that we could spread them out over several months.
However, I had other plans.
No party time tonight.
Mr. Macho decided to take three exams all in one day. At 90 minutes apiece, I was dragging by the time I finished. I did pass for the masters’ level, but it was a challenging morning. After taking
these exams I expect that the CBIP will become a respected certification in the field of data warehousing and business intelligence. However, those who are well read, have sound work experience,
and have a history of professional development should have no problem in attaining this certification.
That was a great conference, rhinestones, champagne, ICCP, and all.