Although some 65 per cent of large European enterprises have adopted a data warehouse strategy, some 60 per cent of such projects fail due to cultural and organisational problems, lack of vision and inadequate technology.
But even if users manage to implement a data warehouse successfully, many find that maintaining, administering and managing it is extremely difficult and costly, which means they end up abandoning their systems. This was the warning from Sean Kelly, managing director of consultancy the Data Warehouse Network, in his keynote speech at the Data Warehousing 97 show this week in Olympia, London.
?Success equates to building a system for the business benefit or value, not thinking about the technology or the performance. Data warehousing is just starting to take off, but when you build one, you need to build it for ever, not just for six to 12 months,? he said.
He cited the most common reasons for failure as organisational opposition to the project for political reasons, particularly when the IT department had not managed to find a corporate sponsor, and lack of vision when the project had not been scoped properly or looked at globally enough.
Other issues included a lack of adequate administration or metadata management tools on the market and the dearth of usable methodologies to act as guidelines when building a data warehouse due to the inability of organisations such as the Metadata Coalition to come up anything.
This situation was not helped by too many users rushing to get involved with the physical design of the system first before they looked at conceptual and logical design to decide what the business need was.
This tended to result in a proliferation of single subject data marts, which were bound to fail because they were too limited in scope to be of any real business use.
To make matters worse, all of these problems are compounded by a chronic shortage of internal or external staff skilled in the data warehousing field.
Engineer calculates that Chengdu's plan to replace streetlights with artificial moonlight would cost $100bn
Dark matter holds the Universe together - and gravitational waves could help identify it
Addison Lee is working on autonomous taxis for commuting and pleasure
IBM and Technical University of Munich team demonstrate how Shor's algorithm, which can't be cracked by conventional computers, can be solved quickly with quantum computing