Better data management is considered a top goal of nearly every organisation big enough to warrant its own IT department. So why isn't anyone actually achieving it?

Australian Graeme Simsion has been a keynote speaker at the Data Management Association (DAMA) International Symposium for four of the past five years, and for all of that time his message has been the same: Data management, at least as traditionally framed, has been largely unsuccessful.

Ever since the concept of data management was first introduced, Simsion says, it has been an enormous source of frustration and failure. The proof is that the "success" stories that are presented to conferences like DAMA are typically more about optimism than genuine success. Simsion has heard countless presenters saying things like: "We've got this function up and running . . . we've got management support . . . it's looking fantastic . . ."

A couple of years down the track it is typically a different story. Most often, it seems, by then the business has lost faith and the function has been closed down.

"A pretty clear indication of a lack of success is if the business doesn't want you, or is not prepared to fund you at the level you think is reasonable or doesn't give you the resources you need in terms of people and so on," Simsion says. "But I think the clearest evidence of lack of success is business actually closing down their data management groups or significantly downsizing them.

"Other evidence of lack of success is that the data managers by their own lights have not achieved the goals that they set out to achieve. So these people are not saying: 'We achieved all the goals and now they're closing us down.' They're saying: 'We couldn't get the support we needed, we struck problems achieving the goals we set out to achieve, and as a result we've lost support.'"

And that is happening all the time, in organisations across Australia and the globe.

Paper vs Practice

A consistent critic of the conventional "data administration" approach, Simsion believes there is a critical need to "re-invent" data management. The business need for data management is not in doubt; what is in doubt is the effectiveness of the methods that professional data managers have used to tackle the problem. Simsion agrees there's nothing wrong with adopting a common approach - as long as it works. Unfortunately, the evidence is overwhelmingly that the data administration approach does not work.

Simsion ought to know. He has been working with organisations on data management and modelling since establishing Australian consultancy Simsion Bowles & Associates in 1982, he is the author of the widely-used text Data Modelling Essentials, he advises the DAMA International Board and he is currently a senior fellow with the University of Melbourne's Department of Information Systems.

Simsion says his experiences working with and talking to literally hundreds of data administration staff in Australia, North America, Asia and Europe over a 15-year period support the finding that data administration is not working. "While some can cite enough achievements to justify their existence, far fewer can say that they had made real progress towards the delivering on the promises which had encouraged the original investment," he says.

Data management is a bit like Communism, in Simsion's view: It sounds utterly sensible and practical on paper, and it is easy to see the problems that arise from not having it, but in practice, generally it has not worked. Data mismanagement is a real issue in most organisations, and one that most organisations struggle with.

Simsion says it has never been particularly difficult to sell the case for better data management. After all, virtually every medium and large organisation - and quite a few small ones - can identify examples of:

• Multiple copies of data with attendant storage, capture, maintenance and translation costs
• Poor data quality, often as a result of holding uncoordinated copies
• Difficulty in pulling data together for operational purposes or to meet management and executive information requirements
• Re-work, as the same database structures and associated maintenance routines are designed over and over again.