- User names
- Fixed values like state or vendor name
- Destination folder paths
Friday, March 25, 2011
Kicking the Share Drive Habit
Business productivity in the world of “just get this thing done already” means using whatever MS Excel can offer to track the content out there and report on it. This process usually equates to lots of manual double checking, lots of verification, and lots of busy work. Lots of human intervention is not a bad thing, however at some point the “finding” and “versioning” become unwieldy. Some type of content management system must be purchased.
The purchasing, deployment, and education of the first system is the most crucial step in setting the stage for the future health of the organization’s information. It was not that long ago when IT Directors were saying that content management systems did not belong in the enterprise services stack. Some industry IT shops still harbor misguided and regrettably wrong impressions about the complexities of information and especially content and process automation.
So here’s a guide to moving content off the share drive and into a content management system.
One of the key concepts that is missed by most ECM vendors is that each bulk import/migration needs to be executed with a certain percentage of customization in order to get the best results. You could bulk import with an off the shelf tool, but don’t expect to get all of your content into the target content management system.
Remember: the cost of exceptions could add up to more than the cost of the off the shelf tool.
Analyze Your Content and Metadata
Use a file listing app (whether it’s off the shelf or home grown) to build inventory Excel/CSV files with the following criteria:
• Absolute file path for folder and file location
• File names
• File properties such as creation date
• Create drop down lists based on the target content management system
• Incorporate the work done in the Analysis
• Design it based on the specific requirements of the import or migration at hand
• Pre-Flight of Excel track files
• Does the file exist?
• Are the date values valid?
• Do User names exist?
• Are Illegal characters handled during the migration?
Make Validation Simple
Using the absolute path values of the source content, make sure the target has an attribute for that same value. This will make is much easier to not only validate the results, but to recover from failures where rerunning thousands of imports (for one failure) would be incredibly inefficient.