The most popular and comprehensive Open Source ECM platform
Let us know more
Data Deluge: 25% of Enterprise Data is Inaccurate
Not only are company data archives overflowing, the data is dirty. Gartner says that 25 percent of Fortune 1000 company critical data collected over the next two years will be inaccurate, incomplete or duplicated. Gartner thinks that three quarters of large enterprises are experiencing dirty data problems and that it won’t be at least until 2010 when they will be able to start getting a handle on the problem. Only three years away, that timeline seems a bit optimistic since dirty data has been a problem that has always plagued systems.
There are many causes of dirty data. For example, the original transaction may have been inaccurately recorded, or the recording of the data might have been delayed. Any time that data gets out of sync with reality, there is a problem.
Dirty data can affect day-to-day operations, slowing down processing. Poor data can also directly affect the quality of the analysis produced by data warehousing projects. In order to comply with regulations like Sarbanes-Oxley and Basel II, accuracy of data is also very important.
What’s the solution? Gartner sees that any solution must approach the problem from a holistic approach. Software tools, like Enterprise Information Management systems, are being made available to address the issues of dirty data. But the solution can’t be software-based only. To fix the problem, the enterprise must adopt a plan that incorporates elements of company vision, strategy, governance, infrastructure, processes and people. Getting the processes and people aligned correctly is the most difficult piece of the puzzle.
Gartner recommends that enterprises set up corporate data management groups. These groups need to act as as ‘data stewarts’ and they need to be trained to manage information as a corporate asset. That would require a big change. Companies that allocate dedicated employees to manage corporate data are very rare. But total commitment from the enterprise is needed in order for the process to work.
