Every summer, I look forward to the TDWI conference in San Diego. The weather is unfailingly warm and sunny. Seaforth, with sailing rentals for mid week breaks, is steps from the hotel. Just beyond
the marina is a park that is great for jogging, playing the didgeridoo, or just watching the world go by. There are good restaurants in the neighborhood, and hotel, the Manchester Grand Hyatt, is
top notch. Of course, the main draw was that TDWI and the major players in business intelligence and data warehousing put on a superior event.
TDWI offers both business and IT professionals courses on everything from hands-on experience with various technologies to workshops on high level management strategies. During the last conference,
I had worked with several products that I had not tried before and attended a BI tools shoot-out. This time I wanted to focus on professional development, management skills, and effective
After a pleasant, early morning flight, I checked in and headed for Mike Lampa’s session titled, “Driving Out BI Requirements Using Group Facilitation Techniques.” Facilitation skills are
something that you never think about until you need them, and, by then, there is usually a lot of blood on the ground. Lampa pointed out that although facilitation does not take the place of a
methodology, it can help teams gather requirements though disciplined workshops. Lampa then went over the roles for facilitation plus some techniques for handling various problems. He stressed that
facilitators need to remain neutral, and that if a facilitator needs to offer advice on the project itself he/she need to step out of their role as facilitator to do so.
Later that evening, I dropped in on Informatica’s hospitality suite. The food and wine were excellent, and Informatica’s presentation was brief, but effective. The presenter made a number of good
points about the value of having a Data Governance Program and suggested that we swing by the Informatica booth later that week to pick up a couple of white papers on the subject. Informatica has
done a good job of creating and marketing their data integration technology. However, they have also made a significant contribution to working with customers to develop effective data governance
policies and procedures in order to get the maximum return for the customers’ investments. Other vendors should be so dialed in to their customers’ needs.
The keynote for the start of the week, Bob Hirschfeld’s “Funny Business: A Satirical Look at Corporate Life,” immediately put the crowd in a good mood. Hirschfeld is a cybersatirist who has
written for some big names such as Jay Leno. Hirschfeld cracked us up by taking the everyday things that we deal with in business and turning the light of basic reason on them, i.e. how do you turn
off your computer? Answer, click the start button. There were painful chuckles when Hirschfeld pointed out the supreme logic of the modern phone tree that finally told him though an automated
response, that he could get the solution to his network problems on the Web. I wish I had a nickel for every time I have run into that very situation.
After experiencing Mark Peco and Dave Wells’ “TDWI Enterprise Business Metrics: Designing Integrated Business Metrics for the Enterprise,” one word imbedded itself in my mind – rigor. This
course took metrics design well beyond that which some might think would be addressed during the course of a normal data modeling exercise. It was designed to fit with the popular management
disciplines such as Balanced Scorecard. In fact, Peco and Wells established the strong connection with business motivation up front, and they pointed out some challenges such as poor leadership,
poor communication, lack of trust and openness, not being action oriented, and poorly defined business goals. However, I couldn’t help but feel that an enterprise that undertook a robust metrics
design initiative such as this class outlined could not help but be the better for it, if for no other reason than the clarity of vision for the business that it would help to bring about.
I decided to close the day with a visit to the Business Objects hospitality suite. The theme for the evening was “Zen and the Art of Enterprise Information Management.” Soft light bathed an
understated décor that favored bamboo. The dinner included sushi. There was no loud, high energy music, no hype. Business Objects staff were on hand to discuss anything your heart desired
from EIM to the sweetness of the dew in the morning. Discreetly behind black panels, masseuses gave neck and back massages to anyone needing to de-stress. I availed myself of their expert hands.
Ommm. Business Objects had redefined the art of the hospitality suite.
Having recently helped a client put a data stewardship program in place, I wanted to compare notes with data stewardship guru, Robert Seiner, so I signed up for his class, “Building Effective Data
Governance and Data Stewardship Programs.” I also figured that this was a good move since I wanted to suck-up to my editor. Anyway, Seiner’s class was great. He was animated, and his approach to
the subject was down to earth. Seiner has created what he calls the 3-D Approach to data stewardship. This copyrighted strategy calls for formalizing accountability for the management of data
- De Facto – giving recognition to the functions that some folks already perform
- Discipline – assigning authority and enforcement responsibilities
- Database – promoting steward relationships to data
I found that his approach rang true with my own experiences. He went on to cover definitions, motivation, and guiding principles for data governance and data stewardship programs. However, the key
point that Seiner made that day was that the return on investment for data governance must be realized through the programs that it supports, such as successful ERP migrations or the delivery of
high quality data warehouses. This connection is well worth remembering for anybody wanting to implement a data governance/ data stewardship program.
After class, I headed down to the vendor show with a several goals. I wanted to verify the star schema tuning parameters for the latest releases of Oracle and Microsoft SQL Server, I wanted to see
the fabled terabyte cube, and I wanted to take a look at a vendor that defied categorization.
The first person that I talked to in the Oracle booth confirmed that yes, the combined foreign keys in the fact table should have a bit-map index. The Microsoft bo0th was likewise quick with the
answer for their product. Put a composite index on the foreign keys in the fact table. It’s always great to come to a conference and talk with vendors that are savvy in data warehousing issues.
I found Unisys camped out next to the Microsoft booth. Not only did they have a large MS Analysis Services cube running on the machine they had in the booth, it was running real-world data from a
national book store chain. It contained masked customer data and it was 2 TB in size! I asked the folks at Unisys to run a couple of queries that I thought might stress the machine a bit, but it
didn’t even seem to breathe hard. Granted that it was a gonzo machine with heaps of processors and an obscene amount of memory, but hey, when you want to work with TB + size cubes, that’s what
you need. It was nice to see that it could be done.
Next, I went looking for a vendor that defied categorization. Noetix fit the bill. Noetix makes products for Oracle ERP customers. One includes a set of views that gives a dimensional look to a
normalized database plus a set of help files that provides standard business names and definitions. Their data warehousing product includes transformations for loading dimensional targets from the
source ERP system. It also includes design guides and the same business metadata that their operation BI offering includes. Although Noetix does sell a query tool, the folks in the booth stressed
that their products are really designed to make the leading BI tools easier to implement and use. Noetix also makes the point that their products are customizable while still providing protection
from major upgrade costs that come with ERP upgrades. Given my own personal experience, I would expect Noetix products to significantly shorten development cycles and greatly reduce maintenance
TDWI conferences are always intense. Attendees can easily be in classes and looking at new technologies for over twelve hours a day. I have developed several techniques for avoiding potential
burn-out at these conferences. In San Diego, the technique is called sailing. In spite of our full schedules, a friend and I managed to get out on the bay for several glorious hours. Seaforth, the
rental company at the marina, had recently updated their fleet of boats, the sun was out, and the wind was just great. We came back charged up and ready to tackle night school. Now is that
dedication or what?
This particular conference had a data warehousing and business intelligence in higher education track. This group seems to have more than its fair share of challenges, so I decided to sit in on
Cary White’s presentation, “Data Warehousing at the University of North Carolina Chapel Hill: A Case Study with a Focus on Lessons Learned.” White gave us a couple of interesting take-aways.
Although getting a good set of requirements upfront is important, maintaining contact with the users throughout a project is just as important. White was candid about the issues that his team
faced, i.e. tight funding and very small software budgets. However, he also gave hope to those institutions with very limited resources, that they could build effective BI infrastructures.
After night school, a group of friends and I headed for San Diego’s Gas Lamp Quarter and Croce’s nightclub. Croce’s offers great food and super music. That night a tight group -including Jorge
Camberos on guitar, Mike Cannon on drums, Max Zape on piano, and Adam Michel on bass – was playing laid back Brazilian Jazz and high energy standards. It was a great night.
TDWI got us fired up for the home stretch with an excellent breakfast, and a solid keynote from Nancy Williams whose topic was “Charting the Path to Real Business Intelligence.” Williams started
out with three rather spectacular BI case studies. It’s always nice to have a good case study in your pocket for discussions with management. These included case 1, ROI of over 1000% – case 2,
reduced operating costs by 50% — and in case 3, ROI of over $6 million. After getting our attention, she went on to say that even if organizations recognize that there are gains to be made through
a sound enterprise BI implementation, many don’t know how to go about putting one together. She went on to describe four stages of BI maturity which went from no data warehouse, no BI experience,
through changed informational paradigms at the functional/departmental level championed by individuals, to the top level where executive management leads the charge to create an enterprise view of
information and its use. Few attendees could have gotten this far in the week without knowing that gaining executive sponsorship is a good thing. However, Williams did a good job of giving us the
case studies and framework to help executives connect the dots.
I had taken TDWI classes from Maureen Clarry before and found them quite valuable, so I signed up for her “BI Manager Toolkit: Managing Change and Productivity.” Clarry wasted no time relating
productivity to anxiety. Not enough or too much anxiety leads to low productivity. She pointed out that there are three things that make employees feel successful – challenge, empowerment, and
doing things that are significant. A sense of entitlement or fear can kill productivity. We went through several exercises that helped us recognize different organizational anxiety levels, talked a
bit about personal accountability, and finished up examining techniques for improving team accountability. If you ever get a chance to take this course, whether you are a manager or not, I would
highly recommend it. Ultimately, we are our own managers.
What a good way to wind up the week! I took Mark Madsen’s “Open Source Adoption in Data Warehousing and Business Intelligence,” thinking that I would get an update on open source technology. I
did get that, but Madsen’s session covered a lot more.
Madsen covered several “open source” business models that vendors have adopted and at the same time pointed out how these models would affect our satisfaction with open source technology. One of
the major BI vendors in this space is offering an open source suite of tools with limited functionality, but if you want the enterprise model, you will have to pay for the pro license. This, of
course, takes you out of the open source arena. However, another open source BI tools vendor offers complete, no-cost, source code. The business model for this vendor includes for fee consulting
and support services. It also includes testing new releases against most other software tools. This second model would appear more likely to be aligned with the services that most open source
The flight out of San Diego was mellow. As we flew up the coast, I started thinking about TDWI in Orlando in November. It promises to be an excellent conference.