# Fractional Beeminding

**URL:** <https://forum.beeminder.com/t/fractional-beeminding/554>\
**Category:** Akrasia\
**Created:** [March 2, 2015, 8:17am UTC](https://forum.beeminder.com/t/fractional-beeminding/554 "2015-03-02T08:17:13Z")\
**Posts on this page:** 8\
**Page:** 1

<div class="post-metadata">

**Author:** ![insti](https://sea1.discourse-cdn.com/flex019/user_avatar/forum.beeminder.com/insti/32/545_2.png) [@insti](https://forum.beeminder.com/u/insti)\
**Post date:** [March 2, 2015, 8:17am UTC](https://forum.beeminder.com/t/fractional-beeminding/554/1 "2015-03-02T08:17:13Z")

</div>

Continuing the discussion from [Beeminding planning](http://forum.beeminder.com/t/beeminding-planning/388):

> [@Beeminding planning](https://forum.beeminder.com/t/beeminding-planning/388/1):
>
> I could commit to the events themselves, but that runs a lot of the same problems talked about [here](http://forum.beeminder.com/t/meta-feature-request-dont-put-yourself-on-the-hook-for-any-more-big-features/). I’d like to make sure that the adventures happen, with sufficient time and planning to make them fun, and I don’t know ahead of time what that’s going to take.
> 
> So I think maybe this idea came from @malcolm, but I’ve started a beeminder for this:  
> [adventure – b – beeminder](http://beeminder.com/b/adventure)  
> I’m allowed to enter any fractional amount under 1 that it seems like the work I’ve done counts for, but I can’t enter the final amount (and actually get to “1”) until travel and accommodations are planned and paid for, and there’s one thing planned for each day of the trip.

I’ve been trying something like this myself, (using percent rather than fractions.) The issue I’ve run into seems to be that the fraction that I report on any day seems to be heavily influenced by how much I need to do to not derail, rather than how much of the task I feel I have completed.

I fear this is leaving me with hidden road debt, as my road makes me think I’m further along than I am and I will end up with a lot to do in the final days.

It might just be a result of my task being relatively small and my timescale being short.

I’ll post another update when I’ve completed the goal and seen how things turned out, but though I should post my ‘in progress’ concerns in case anyone else has the same worries in the future.

---

<div class="post-metadata">

**Author:** ![drmaciver](https://sea1.discourse-cdn.com/flex019/user_avatar/forum.beeminder.com/drmaciver/32/80_2.png) [@drmaciver](https://forum.beeminder.com/u/drmaciver)\
**Post date:** [March 2, 2015, 9:29am UTC](https://forum.beeminder.com/t/fractional-beeminding/554/2 "2015-03-02T09:29:52Z")

</div>

I think this record might be broken, but let me sing the song of my people:

Beemind time spent on a task, not output achieved. It’s harder to inadvertently lie about and means that your limiting factor is akrasia, not planning failures.

---

<div class="post-metadata">

**Author:** ![byorgey](https://sea1.discourse-cdn.com/flex019/user_avatar/forum.beeminder.com/byorgey/32/70_2.png) [@byorgey](https://forum.beeminder.com/u/byorgey)\
**Post date:** [March 2, 2015, 2:08pm UTC](https://forum.beeminder.com/t/fractional-beeminding/554/3 "2015-03-02T14:08:56Z")

</div>

I mostly agree, but as a counterpoint, I do sometimes find it useful to beemind output achieved, especially in cases where my perfectionism can cause me to spend way too much time on something without actually finishing anything. For example, this worked well when I was writing my dissertation proposal: I had two goals, one for time spent and one for pages written. The pages goal was modest but often forced me to get creative about incorporating text I had previously written (rather than rewriting from scratch) and/or just get down to the business of writing rather than doing things like fiddling with the typography or spending a long time making a single figure.

Put another way, not all time spent on a task is equal (and this goes more for some tasks than others). Adding some modest output tracking helps make sure I spend my time in high-impact ways.

I do agree, however, that if you do beemind outputs it works much better to beemind objectively measurable ones (such as page count, blog posts written, phone calls made, etc.) rather than more subjective ones such as completion percentage, so that it is less tempting to lie about the numbers.

---

<div class="post-metadata">

**Author:** ![dreev](https://sea1.discourse-cdn.com/flex019/user_avatar/forum.beeminder.com/dreev/32/4311_2.png) [@dreev](https://forum.beeminder.com/u/dreev)\
**Post date:** [March 2, 2015, 5:49pm UTC](https://forum.beeminder.com/t/fractional-beeminding/554/4 "2015-03-02T17:49:30Z")

</div>

I think @bee’s and @malcolm’s idea has been that you’re actually allowed to make up any number you like for the intermediate datapoints. But reaching 1 or 100% is wholly objective. So as @insti points out you can paint yourself into a bit of a corner, but that danger is incentive enough to be realistic about the fractions you report. Maybe this works best, or is only fully robust to extreme akrasia, if you can always get from near 0 to 100% with a single all-nighter if push comes to shove.

I agree with @drmaciver about [preferring actions over outcomes](http://blog.beeminder.com/whattomind "The 3 tips for picking a metric: 1. Prefer things you can automate; 2. Prefer things that make interesting graphs (QS First principle); 3. Prefer actions over outcomes") but I like beeminding outcomes too, just more conservatively, as @byorgey says.

---

<div class="post-metadata">

**Author:** ![drmaciver](https://sea1.discourse-cdn.com/flex019/user_avatar/forum.beeminder.com/drmaciver/32/80_2.png) [@drmaciver](https://forum.beeminder.com/u/drmaciver)\
**Post date:** [March 2, 2015, 6:03pm UTC](https://forum.beeminder.com/t/fractional-beeminding/554/5 "2015-03-02T18:03:10Z")

</div>

Yeah, TBF I’m not actually quite as religious about beeminding actions over outcomes as I claim to be. I’ve got some goals which are outcome based, they’re just outcomes which map very linearly to actions - it’s unlikely that I can ever be in an eep day that I can’t solve by just putting in a bit more effort than I normally would (word count on blogs, number of exercises done, writing a particular email once every 3 weeks, etc). I guess it’s more that I like to avoid goals with unknowns in them.

---

<div class="post-metadata">

**Author:** ![tomjen](https://avatars.discourse-cdn.com/v4/letter/t/839c29/32.png) [@tomjen](https://forum.beeminder.com/u/tomjen)\
**Post date:** [March 2, 2015, 7:50pm UTC](https://forum.beeminder.com/t/fractional-beeminding/554/6 "2015-03-02T19:50:25Z")

</div>

Another vote for actions over outcome. If I had beeminded my Udacity goal in lessons rather than hours I would have been SOL this weekend when the work I had to do for one lesson suddenly took 2-3 times as long as they usually do.

Now? It just takes me a little more time to work through it, no issue.

---

<div class="post-metadata">

**Author:** ![mbork](https://avatars.discourse-cdn.com/v4/letter/m/ed655f/32.png) [@mbork](https://forum.beeminder.com/u/mbork)\
**Post date:** [July 31, 2021, 7:05pm UTC](https://forum.beeminder.com/t/fractional-beeminding/554/7 "2021-07-31T19:05:07Z")

</div>

Hi there,

it just occurred to me yesterday that I use a similar semi-cheating technique which sometimes helps me ease things a bit short-term, but is 100% objective long-term. I have quite a few “do-this-almost-every-day” goals with rate of 0.8 “times per day” and a buffer cap of 2 days. For example, one of these goals is putting the expenses into a ledger file. I usually do it in the evening, but sometimes I know this is going to be a lot of work, so after my groceries round in the morning I enter this one expense and mark this goal as done for that day. This way, I have a “day almost off” essentially for free, but I know that I’ll need to make up for it the next day.

Makes sense?

---

<div class="post-metadata">

**Author:** ![insti](https://sea1.discourse-cdn.com/flex019/user_avatar/forum.beeminder.com/insti/32/545_2.png) [@insti](https://forum.beeminder.com/u/insti)\
**Post date:** [January 4, 2022, 6:54am UTC](https://forum.beeminder.com/t/fractional-beeminding/554/8 "2022-01-04T06:54:03Z")

</div>

> [@insti](#):
>
> I’ll post another update when I’ve completed the goal and seen how things turned out, but though I should post my ‘in progress’ concerns in case anyone else has the same worries in the future.

It looks like I forgot to do this, and since I’ve forgotten which goal I was talking about I’m not sure I can post an update.

> [@insti](#):
>
> I fear this is leaving me with hidden road debt, as my road makes me think I’m further along than I am and I will end up with a lot to do in the final days.

I suspect this was correct, but having the goal meant that I’d done _something_ on the goal and so was further along than if I’d just procrastinated up until the night before the deadline, even if the “percent done” value did not match the percent done of the task.
