There’s a lot of interest in our experiences delivering data governance capabilites. But that’s not because our clients want to see us huddled in their hallways on Monday morning, our 1-pump double-shot Pumpkin Spice Lattes half gone (it is fall now, after all, and the frappuccinos are SO five weeks ago!), our laptops collectively warming up in anticipation of another fun-filled data governance workshop.

While I have a few different theories about the traction our data governance services are getting I believe the main reason is this: people—usually IT people—realize that in order to get business buy-in, they need to establish a common vocabulary around data governance. The kickoff-and-cold cuts approach to data governance that I’ve discouraged in the past doesn’t give the IT visionaries either the credibility (“Why are we in another meeting about data?”) or the organizational authority (“I’m not participating in this unless someone tells me I have to”) that it needs to engage business people in a sustained dialog about managing data as a corporate asset.

Indeed the common vocabulary is critical. And I don’t mean explaining to end users for the ninth time why metadata is important or catching an executive in the break room long enough to expound on how the semantic web will change life as we know it. I mean introducing the concept of a data trustee council, discussing how decision rights work and why data quality should be measured according to business benefits.

As with any initiative that involves business and IT alignment (and data governance is by its definition business-driven and IT-enabled), introducing data governance into an organization requires missionary work. Maybe your internal team can begin proselytizing data governance. Maybe you can even link it to an active and high-profile initiative in your firm. Maybe your business users, driven by a little curiosity and perhaps even a double-shot latte, could simply use a nudge.

Technorati tags: data governance, data management, data governance workshop

So I’m at the Teradata Partners conference last week in Vegas and, after the requisite conference theatrics—I’ve come to expect hydraulics at any vendor event with over 3000 attendees—down comes Mike Koehler, he having recently taken the helm of the newly-independent Teradata, talking about his company’s plans for the coming year. Then there’s the inevitable dry ice sublimation—I’ve come to expect dry ice at any vendor event with over 3000 attendees—through which appears Dr. Jim Goodnight, he the founder and leader of the venerable SAS. What gives?

What gives, indeed. The two companies have sometimes been at odds. Marketing product managers at a large bank we work with insist that they can do everything they need to do in SAS tables, extracting their data from the Teradata system and manipulating it in SAS datasets—and not enriching the data warehouse with the results. The product managers wanted their own copies of the 8-terabyte data set, essentially putting the kibosh on the bank’s goal of “single version of the truth.”

SAS intellectual capital was tied up in its tools, Teradata’s in its database. Now the two companies have announced a formal partnership. There’s a peanut butter cup analogy in here someplace.

Teradata Partners was abuzz with speculation about the partnership. What was in it for SAS? “Better performance, for one thing,” one smart SAS rep told me. High-speed, in stream analytical applications seemed to be the theme. On the Teradata side it was more about the application suite. “We’ve relied on a range of partnerships to provide a pretty broad set of functions,” a Teradata manager explained. “SAS offers a suite of tools beyond statistical analysis.”

Soon SAS analytics will run locally on the Teradata platform. It appears that the might of SAS has convinced Teradata to support non-SQL based processing, and non-Teradata database functions being run inside Teradata’s engine. SAS processing can in turn capitalize on the performance advantages of Teradata.

But in the longer term, the dark horse might turn out to be data quality. Historically Teradata has relied on its customers having an incumbent data quality solution, essentially bolted-on to the Teradata load utilities, While not yet part of the two companies’ formal plan, embracing SAS progeny DataFlux and its data cleansing and matching capabilities might well be the partnership’s ace-in-the-hole.

Technorati tags: SAS, Teradata, SAS and Teradata, embedded analytics

Whenever my schedule permits I like to host podcasts for our friends at BI Network. I’ve talked with many of the network’s vendors and sponsors on the record. (I’m extremely ambivalent about the fact that I sound like my sister on podcasts. Heredity’s a sunovabich.) Anyway at the last TDWI conference I hosted podcasts with Microsoft, Teleran, Syncsort, and data warehouse appliance vendors Paraccel and DATAllegro.

I'm excited about the data warehouse market in general, and DATAllegro was announcing its new grid-enabled data warehouse appliance at the conference, and had a cowboy-themed booth with a John Wayne mug shot and the requisite Western garb. In discussing his company’s latest product announcement, DATAllegro CEO Stuart Frost said something intriguing. He claimed that appliance technologies can actually encourage business-IT alignment.

Given that I teach a full-day course on business-IT alignment for BI, in which I claim that the processing platform doesn’t matter, I leaned forward in my chair and my chin hit the mike, so when you hear a little “bump” in the podcast, you’ll know where it came from.

"New grid-enabled DW appliance technologies enable large corporations to build highly scalable hub-and-spoke data warehouse architectures,” Stuart explained. “What this means is that business units can purchase independent data mart appliances (spokes) to meet their own budgets and SLAs, while central IT can take care of corporate data governance and management on an enterprise-class hub appliance.” (Stuart was dressed as a cowboy when he said this, but the fact that he was implying specific organizational ownership delineation as a result of data warehouse appliance adoption kept my attention focused.)

“This 'best of both worlds' approach decreases the tension between business units and corporate IT that can often arise with traditional centralized data warehouse technologies." Hmm. Business accountability of data in the context of business requirements. Sounds like a familiar admonishment by yours truly. Perhaps the data warehouse appliance is a cheaper, faster way to not only process and deploy larger data volumes, but deliver data to different constituencies based on ownership and usage context. In which case, Yee-haw!

Technorati tags: data warehouse applicances, DATAllegro, BI from Both Sides, business-IT alignment

In 1985 I went to work for a technology start up. I’d left a job with a well-established high-tech firm to take the plunge with a fledgling database machine company that was doing innovative work with massively parallel technology. Friends called me crazy. All I knew is that I’d left a big company, agreeing to take less money and more stock options from a tiny firm started by a guy in a Santa Monica garage. I think I threw up when I signed the offer letter.

The mid-1980s at Teradata was a heady time. There were 12-hour days, beer, doorway design sessions, parking lot volleyball, beer, moment’s-notice travel assignments, paintball, beer, and an esprit de corps that transcended culture, cementing the company’s very brand. Teradata developers, far from being hidden in the subterranean alleyways of headquarters, were revered and executives let the light shine brightly on them. Between executive briefings with the likes of Citibank and AT&T—enterprises beleaguered by large transaction volumes and struggling to understand their businesses—we’d go on midnight grunion runs at Playa del Rey, then back to work to fling a little more code or finish the manual, then reconvene to devour sweet shrimp (with heads) at O-Sho. Hard work was in our DNA. Most of us were young, cutting our teeth on cutting edge technology by day, and partying like rock stars (albeit poor ones) by night. We merged, de-coupled, re-normalized, and disaggregated—and I’m not talking about data.

Because of Teradata I designed one of the world’s first terrorism databases, took a coma query from minutes to seconds, christened my favorite London pub, deconstructed balance sheets, learned the words to "Rocky Top," built data models sans dangling foreign keys, learned to love—and then hate—Sambuca, saw the High Atlas mountains at twilight, and delivered customer profiles before they were a gleam in Tom Siebel’s eye. Because of Teradata I forged permanent friendships, sat at the feet of the masters, and met my mate.

I left Teradata in 1991, before the AT&T acquisition that eventually made Teradata a subsidiary of NCR. When I left, Teradata was still independent. This week Teradata is independent once more. The stock ticker is TDC. Congratulations to the new company, and to the old company, and to the pioneers who built them both.

Technorati tag: Teradata, Teradata spinoff, data warehouse pioneer