I think I’m a relatively typical Beeminder user. I investigated Beeminder a few years ago, and started using Beeminder daily in February for weight loss. I deliberately didn’t add new goals for a while, because I would naturally add too many goals and be overwhelmed and get demotivated. I added a few a month or two later, and in June or July, I made a big list. I added most of the goals on my list immediately, but I left a few harder ones on the list.
I realized yesterday I still had left them on my list, not added to Beeminder, not being worked on at all. So, I added one, determined to get started, but I almost immediately realized I’m going on a trip next weekend, and “I shouldn’t start before then, I don’t want to immediately fail.”
I had an epiphany, and I’m not sure if it’ll bear out in practice, but here it is: I added the goal, and immediately did Take a Break and gave myself another flat week. I don’t need to make progress on the goal this week, but I have a stronger commitment to actually start after I get back.
I’ve done that, and it worked great. I think Beeminder gives a wonderful twist to the old problem that deciding you are going to start doing something in a few weeks is almost counter productive, since you feel like you’ve accomplished something by making that resolution but really haven’t done anything.
The only downside I see is that you lose the ability to realize you have set it up wrong the first week and immediately cancel it.
I often go a step further and make the goal rate 0 until I have enough data / determination to commit to something. I know opinions differ on this, but effectively this is using Beeminder as a Quantified Self engine without a commitment contract. Then when you’re ready to actually commit it’s super easy.
I’ve got a goal for working on preparing my tax return that is flat for another 3 months before it starts sloping upward. It’s much easier to commit your future self to doing something and the longer away it is the easier it is to commit to it!
Don’t forget about @mary’s autodialer script that gently increases the goal’s slope based on actual doing. You may still need a bump to get you started, of course, but it should reduce the danger of overcommitting at the outset.
(I like the idea of autodialing so much that I think it should be a standard feature, to reduce goal-creation friction.)
I can definitely see autodialing being important to some people.
I have “review my beeminder goals and progress, and adjust buffer and rate” in my weekly cadence, and it seems to be working well since Feb+, except for the evergrowing “I should someday Beemind this” list. (which I hope this fixes.)
One of the things that is highlighted for me when reading the Beeminder forums is how we are all broken in different ways
For not making too many goals, I was recently really taken and inspired by this BM blog post. Basically the principle is that you can’t add any new goals (at least, not active ones – so zero-rate goals don’t count) until all your bees are green. I am a serial add-too-many-goals user, but this way I’m restrained.
Give it a try. It really worked for me after years of always overestimating how much I could follow at one time.
Checking back in here a year later. Beeminder now lets you add a bunch of safety buffer when creating a new goal, so you can do this super easily! Just make a new goal, set your desired rate, click the “extra leeway safety buffer” checkbox, and put in how much margin you want!
What helps me is always having a matrix of Beeminder goal widgets on my phone. For many years it was 9 goals. Now it’s 12. Whenever I complete a goal successfully or decide to retire one for any reason, I know I’ve got to figure out another one soon or face an unsightly blank space in that matrix. It really helps to keep me honest!
I have had the opposite problem where I create a lot of poorly thought out Beeminder goals on impulse.
The biggest complication comes when the new goals overwhelm my notification system. For example, Android notifications only show the top few items, so if I’m on the go on relatively busy days, I might not notice some of the goals.
The Complice integration helps with this, since it helps me incorporate what’s due for the day in my planning. Not applicable to non-Complice users, though.
I think the 30 days is an interesting magic number for safety buffer. I haven’t trialed a Beeminder goal with a safety buffer that large probably because I have been anchored to the 7 day default. So I’ll give it a try.
Hey, if this is a best practice people should know about, why not make it part of the interface for a more guided goal creation? Maybe the 7 day safety buffer is appropriate for most cases (do you have data on this?) but the new people who are trialing Beeminder goals are people who probably need a lot of guidance anyway. Down the road sort of thing, but in my very unqualified, don’t know squat about UX opinion, clearer built-in onboarding could get a lot of new people to try it out further.
Daniel asked me in an email to chime in here. (Well, that was almost 2 months ago, but still.) Here’s my take on this: I’d start the goal with next-to-zero buffer (say, 1-2 days), but with the rate equal to half of what I expect to be able to do. Then maybe dial that up after a week or so.
