Without trying to shove David Allen's Getting Things Done philosophy down anyone's throat, one of the very good points he makes is that you should plan tasks, not projects. For instance, a task item called "clean garage" isn't nearly as good as a set of tasks like "clear off work bench", "bring old clothes in garage to charity", and "throw out old holiday decorations".
The trouble is you need to stop and make time to think out your projects. While this takes a little brain time up front, it can save you a lot of labor down the road. The trick is avoiding the convenience of cheating. I've caught myself doing it several times with entries like "increase involvement at kid's school" or "settle Jones case". These aren't tasks; they are projects. The trouble is, when the idea occurred to me to do these projects, I didn't have time to break them down in component parts so I just put a useless project name in my task list. When I run into such vague generalities while executing my task list, I just shrug my shoulders and push the project off to another day.
Realizing this habit was getting me nowhere, a few months ago I began a new practice. I "plan to plan." So instead of listing the project as the task, I now make a task item to plan a project.
While I may not have time to initially plan the project, I can task myself to do it later. When I have time, I plan the project out and get the satisfaction of ticking off the original "plan project" entry. This way, my task list is not cluttered with white noise but actual items I can accomplish.
An additional use for this hack is when you anticipate a fork in the road of a project. For instance, I often plan projects for my clients where I know steps 1 through 5, but step 6 is contingent upon what happen between steps 1 through 5. In that case, I enter a task as step 6 called "Plan project further." Again, when that task shows up I will have the information I need to plan further and can do so accordingly.
For instance, in the above example, the other guy may simply accept my contract, have revisions, or tell me to pound sand. I don't know what will happen when I start the project, but I will know by the time that task shows up. Then I can plan further as required.
What are your ideas for tackling this problem? Drop me a note or leave a comment.