“80% of success is showing up.” — Woody Allen
“It should be completely implausible to describe a startup’s CEO as a flake.” — Paul Graham and Jessica Livingston’s heuristic for successful startups
“Let your ‘yes’ mean yes, and your ‘no’ mean no. Anything more than this comes from the evil one.” — Matthew 5:37
For the past two weeks I’ve been trying a new system I’ve formulated to make all casual commitments be hard commitments. It’s similar to my Anti-Resolution Resolution, which has worked well for me for years now. I publicly committed to never musing that “I should” do something without taking at least some small action to make that thing more likely to actually happen.
Now I’m trying something similar for “I will”. Specifically, I can never just say, no matter how casually, “I’ll X the Y”. (File the report, call the dentist, send the pictures, etc.) I have to add consequences if the Y ends up un-X’d.
I should admit that the Anti-Resolution Resolution has mostly just made me not say “I should” in the first place. It hasn’t been life-altering or anything. But it’s worthwhile in that I feel less guilty about not doing things I “should” do. At worst the “no consequence-free ‘I will’” system will be similar. But I predict it will be much better because of what it means for your relationships with people when you’re that conscientious. So far it feels quite good.
The pragmatic question was how to commit to consequences on the fly without bogging down everyday conversations. My first ideas were putting tasks in GTBee, a Slack bot that mediates conversations, or a separate Google calendar for “I will” commitments that’s integrated with Beeminder somehow.
But, by the Shirk & Turk principle, I’ve started with something I can do manually. Namely, whenever I make an “I will” statement (even the most casual “cool, I’ll take a look” type utterances) I explicitly also mention a deadline and put it on my calendar and on a Beeminder goal. It feels strange to enter datapoints in the future on Beeminder but it’s always been allowed. I make the datapoint value be 0 until I actually do the thing, when I can change it to a 1.
Beeminder isn’t enforcing any particular success rate here, just that I keep 8 commitments per week. But my success rate for my first 22 “I will” commitments has been essentially 100%. I say essentially because the other day I was 10 minutes late on one of them. Which might be a dangerously slippery slope because on the very next one I was 20 minutes late. We’ll see!
I’ll be sure to update this post to let you know if this system stands the test of time [dreev.es/will deadline = Oct 27 5pm].
Late Penalties
UPDATE after another 2 weeks: I’m continuing to hone the system and it feels kind of amazing. I came up with a way to deal with the case of missing the deadline but still meeting the spirit of the commitment. The idea is to count it as partial success, so you can still compute an overall success rate. Here’s what I’m going with:
- by the deadline = 100% success = datapoint of 1 in Beeminder
- seconds late = close enough, w/in measurement error, still 1
- minutes late = .999 (baaasically counts)
- hours late = .99 (no big deal, almost fully counts)
- days late = .9 (hey, it got done, that was the main thing)
- weeks late = .5 (kind of half defeats the point to be this late)
- months late = .1 (mostly doesn’t count)
- years late = .01 (better late than never, barely)
So with 47 I-wills and counting, I’m holding steady at greater than 99.9% success.
UPDATE 2017-10-27 5:11pm: Going on 3 months, I’m still averaging 8 commitments per week, at 99.97% reliability. (I’m some minutes late on my commitment above to make this update so I’m getting 99.9% credit for this!) So I’m absolutely convinced this needs to be turned into a system anyone can use and I’m committed to finish implementing it in the next couple months! (dreev.commits.to/finish_implementing_this_system/by/january!) Of course with over 100 commitments and counting, my overall reliability will still be over 99% even if fail to do that.
UPDATE 2018-08-29: It’s been over a year and I still love this. I’ve now made 379 commitments and have an overall reliability of 97.25%. We’re developing Commits.to, as it’s now called, very gradually and it’s still in private beta but feel free to get on the waitlist for when we eventually open it up. (Or commit to contributing to the open source project to skip the queue!)
Thanks to Bethany Soule, Laurie Reeves, and the daily beemail subscribers for reading drafts of this.