First, training (or learning) cannot be a static thing. If you take an excellent trainer and a student on a good day, the student may remember 70% of what is presented in a class. I am still learning new things about Worldox and other programs. So learning must be ongoing. Some things that help in this area are brief on-demand videos or other resources (Worldox has about 50 3-minute videos of specific functions at www.worldox.com/howtovideos). Similarly, “brown bag lunches” where the firm pays for lunch and specific features are presented and questions answered can be useful. I have also had good luck with distributing specific “tips” twice a week or so by email. Users can absorb one specific item, and also get the sense that their input is encouraged.
Secondly, training should aim at getting people to actually understand how a program works, not just using specific features by rote. To take an admittedly difficult example: someone may be taught how to use specific Word styles for various functions, but still have no clue how to create even the simplest style. Similarly, when installing Worldox recently in a new firm that had broken away from a previous firm, one user said: “I used to have a button that did X.” Much better that user should learn how to create bookmarks rather than just clicking on a pre-set option. This is the technology version of the “give a person a fish vs. teach them how to fish” paradigm.
In a perhaps futile effort to get users to RTFM, I use pamphlet-size “cheat sheets” for training instead of full-size manuals (which tend to live underneath the desk under the extra pair of shoes).
The bottom line is that poorly trained users cost a firm money. And without ongoing effort to fight entropy (the tendency of any system or body of knowledge to deteriorate without outside input), the standard level of knowledge will deteriorate. In the KIA example, KIA demanded a 15% fee reduction because firms were wasting time and money using poorly trained lawyers.