I seem to be moving in the opposite direction, from bucket goals to specific sub-goals.
Some of my bucket goals are automatically updated by other goals. Exercise is a good example; it can be a ‘binary’ goal that gets a point if one or more of my gym, runkeeper, fitbit, or other goals are updated that day.
I’ve oscillated on this, with bundling and unbundling of my goals seeming to be the right thing for me to do at different points in time. It might depend on the kind of ‘stuckness’ that I’m struggling with.
I think part of what drives the bucketing is that most of the things that I’m currently interested in beeminding have no feasible associated autodata, so managing a lot of goals becomes comparatively annoying.
It’s also that my beeminding seems to naturally go in expand / contract cycles. Bucketing things together as my goals contract seems to help counteract that, so I’m still keeping reasonable control using Beeminder despite the fact that I’m down to two goals (well, three but one of them is heading towards archive).
One nice side-effect of bucketing is that it lets different elements trade off against each other - the question often becomes not “Do I do this goal or skip out on it?” but “Which of the various ways to satisfy this goal do I not want to do least?”
Mine too. I’m getting better at spinning up new goals to support areas that need it, or shutting down goals that aren’t serving me very well any more.
Benefit and curse. I don’t always have the presence of mind to trade off effectively. And some possibilities (that I would like to do) face starvation. At that point I spin up a dedicated goal, and the bucket becomes less bucket-like.
If you have a dedicated goal for each task, plus a bucket goal (autodata from the task goals) you can guarantee some minimum progress on every task while requiring that you do more than the sum of the minimums on the bucket goal. Not sure if that is too much overhead though.