The literature of our industry is replete with advice about why data warehouses fail and what makes data warehouses successful. One topic that gets only cursory coverage, however, is what to do when they are in the process of failing. In my own practice, this is a pretty robust business; and I initially wanted to evoke the image of Red Adair and his gang rushing off to do the impossible extinguish oil well fires. Some more thoughtful marketing minds prevailed and I dropped it; but, in a way, data warehouse rescue is very much like the Red Adair act. Projects are caught in limbo between success and failure, and the precise steps needed to save them are not clear to the people involved. However, one principle stands out above all others: Rescue efforts are not the place for revisioning or going back to square one. If rescue is called for, the only appropriate thing to do is to get things back on track or, in some cases, write the whole thing off. Not every failing project can be saved.
Failure rates of data warehouse projects have been discussed for years. One study claimed it to be more than half, perhaps as much as 80 to 90 percent. Two columnists in this magazine had a heated debate over those numbers, with one quoting his own survey pegging the number at five percent. Such a discrepancy just proves Mark Twain's point that "There are lies, damned lies and statistics." Whatever the number, I personally believe that outright failures of data warehouses are far less common today; but, by the same token, achieving an unqualified success is just as uncommon. Between the rescue efforts I've been involved with and those I'm aware of, two broad categories of missteps that lead to data warehouse purgatory are apparent.
Register or login for access to this item and much more
All Information Management content is archived after seven days.
Community members receive:
- All recent and archived articles
- Conference offers and updates
- A full menu of enewsletter options
- Web seminars, white papers, ebooks
Already have an account? Log In
Don't have an account? Register for Free Unlimited Access