One of the biggest impacts on the end-user when implementing an ECM system is the time it takes for end-users to enter metadata. A user would state this issue as: The way the ECM system was designed requires that I enter in many metadata fields before I can store information in it. While these fields may be important to the company, I don't have time to fill them out.

The user in this case will try to find some other way to store their information, in some other repository, resulting in a system that is seldom used and lost shared company information. In reality, the user would not like to enter in any metadata while the company typically requires numerous tags. Well intentioned ECM designers want to give the business managers all the metadata they ask for, but the business managers seldom understand the impact of the effort required by their employees. Practically speaking, it is necessary to provide a balance that will satisfy both the end-user, because they need to use the system, and the company, because it needs to adequately organize and maintain the information.

Balancing the metadata requirements of an ECM system is best accomplished when designing the system by determining the "user-effort-threshold" for the type of information being stored. How much effort an end-user must output to store information must be justified by risk to the organization. For example, if the end-user is storing information which will pose a risk to the company if it is not managed correctly (e.g. regulatory related documents), then an end-user will better understand the need to put more effort into metadata entry. On the other hand, if there is little risk to the information being stored, then the end-user will not want to make the extra effort. In this case, it would be best to have the taxonomy of the ECM system provide default metadata inherited from where the information is placed in the system. If the taxonomy is done correctly, this will provide the basic tags the company needs to organize and maintain the information while minimizing end-user effort and increasing end-user acceptance.