I wanted to emphasize one of the side-points Tetsujin made above. You really do have to examine yourself to figure out what system is going to work for you. When I was first starting out, I was struggling to find a system and I kept reading that same advice. I never followed it, though, because I'm not the type to sit down and contimplate myself or meditate on an issue and get anything productive out of it. I found a more practical and hands-on way to approach pretty much the same goal: simplification.

(And, as a simplification of this post, I'll let everyone know that the next few paragraphs are my personal experiences and the second-to-last summarizes it all into more generalized conclusions, in case you don't care exactly what I did.)

For the first few months that I was following GTD, I never did a weekly review. I just relied on my transfers between systems that I did every week or two. I'd become dissatisfied with some litle aspect or I'd see a new shiny toy and I'd switch. I never stuck with one system long enough to need a review. I tried computer-based systems, but spent too much time fiddling with them and optimizing them and not enough time doing things. Also, I'm not next to the computer all the time, so it required printing things out every day to be effective, and I coulnd't be relied on to do the printing. So I went to paper-based and I simplified. A lot.

When I first switched to paper, I had index cards and a Moleskine calendar. That's all. My hard-landscape was in the calendar and kept one index card per context with a list of NAs on it. As I found things that I needed, I added on piecemeal. I needed a project list--I got a Moleskine cahier; I needed more in-depth tracking for homework (I'm in college)--I added another stack of index cards with all my assignments, one per card. I didn't let myself change the system. I just found ways to fit in new information as I needed it. I started doing reviews, and found that what really works for me is more frequent, smaller reviews. In the down-time between classes I'd flip through my project list and my index cards and make sure everything was moving along. I didn't do big formal reviews, just small ones as something started nagging me or a had some free time.

After a couple months, my system was stabilized. So I started thinking about what I might need to change about the system to make it really work. Only after I'd figured out everything I needed out of a system did I let myself change the base system. And at that point, the solution was obvious for me. I need my planning to be on paper and I need to always have it with me. The one nagging problem I had was that there were too many individual pieces. So I bought a Filofax and I haven't looked back. It's working perfectly. I steered clear of the DIY planner route because there's so much potential for endless tweaking down that road. I limited myself to stock refills--the only ones I use are week-at-a-view calendar, to-do list, lined paper, and address book. You can put almost anything on lined paper.

So essentially, what I'm saying here is that , by drastically simplifying and adding back in only what you need, you can find the habits that work for you without sitting down and trying to pull it out of thin air. At the end of the process, pick the one big problem you can't solve by adding on and solve it. For me, I needed everything in one place, so a planner binder worked. For you, maybe you need the connections between projects and NAs to be maintained for you, so you'll switch to a computer. Or you'll have some other problem to solve. But I definitely think that working up from radical simplicity is the way to figure out what you really need.

Please excuse my long-windedness and I hope this helps you, or at least gives you something good to think about.