Specific day of the week goals

There a lot of things in life that need to be done, but only on, say, Tuesdays or Sundays. There is no way to set up something like that using Beeminder without setting up a bunch of things exactly right, i.e. derail behavior/deadlines/when it starts.

Like, if I derail on a Tuesday (t=0) goal, I would like it so that it makes me now do it on Wednesday (t+1). But for the next week, the eep day should be on a Tuesday (t+7) still.

You have to pay other people (IFTTT/Zapier) to let you do something that IMHO Beeminder should have always had. And I’m not even sure how you’d do that.

I’ve been trying to search through this forum just to make sure this isn’t something that people are always complaining about or something with an easy fix, and the only things I have been able to find are the below:


They don’t really address the issue that I have, namely, the difficulty in setting them up correctly, making sure when you derail if it gets back to the right spot, etc.


I’m not sure why I didn’t reference beescheduler

For those that use it, I’d like to point out that it doesn’t play well with scheduling breaks (doesn’t work at all), and I don’t think there is any active development. The guy who made it @enolan hasn’t been active on this forum since 2017.

There are so many instances where this feature is much needed for myself personally:

  • accounting for weeknight goals, like anything related to bedtime routines, or planning for the next day
  • workouts: if I miss a day, it may not be feasible to then do 2 in a row, or move the entire schedule over to accommodate. .
  • deep work. It would be much better if I spent 1h of SQL interview practice problems MoWe, and 1h applying to knowledge to a project on TuTh. Instead, I end up doing 1/2 an hour on each, which significantly erodes the quality of the work I do.
  • Fridays off - Frankly, I don’t really need to be doing anything on Friday, and I actually often start my work on Sundays, depending on football season.

Really, beeminder doesn’t work for anybody who’s not a M-F worker. I understand that probably covers 98% of users. My guess is that’s why this issue doesn’t get more attention.

But if there is already a “weekends off” feature, why can’t there be a thursdays off feature? It’s like if beeminder only supported EST

…but this is an extremely basic feature to have. It is crazy to me that the devs don’t wake up every morning thinking about this.

There is not a task management platform, habit tracker, alarm clock, or calendar system with recurring events in existence that does not have the ability to do this.

The response from beeminder on the forums has been a polite:

What I’ve learned from support has given me the impression that this isn’t in the pipeline, so I’m going to tag @dreev here.

This post also gets at the issue.


Thanks for bumping this and talking about what you’re interested in seeing! It’s always helpful to know what really bugs people and what they feel is lacking.

So, I might have said this before via email, and I don’t actually make the decisions for what we do in terms of development, but I have some thoughts about why we don’t wake up every morning thinking about this that might kind of put things into context. One answer is that we get very few requests for this kind of thing. I track all our feedback, and it doesn’t even hit the top 100. It looks like we have 6 votes for it since I started my spreadsheet system in 2019.

Now, the powers that be don’t only make decisions based on the raw number of feedback requests we get for something, but getting so very few requests for it is a pretty compelling argument that it’s not that critical for most users.

Another answer is that Beeminder is totally agnostic to when you do tasks, and always has been, and that’s baked in. It’s just a straight-line graph, after all, and Beeminder doesn’t care how much you do or when as long as you’re on the right side of that line. It’s difficult to think how we could make that do what you want and make things due on specific dates without ending up with a gazillion road segments, which isn’t great. Even weekends off cause some problems there. (From a support point of view on that, I dread to think what fixing people’s graphs would be like if I have to take into account a bunch of custom parameters about what days things should be due – it would be a nightmare, quite frankly.) I don’t think that the way things work behind the scenes is that compatible with the change you want.

That said, I’m not a developer and maybe there is some plan behind the scenes that we could eventually implement – and I’m not saying for sure that it’ll never happen, or when it might happen if it does. People who need this kind of thing really seem to need it, which is useful to keep in mind, and hearing about people’s use cases for this is always helpful. But I hope this kind of answers why I think it hasn’t been a priority up to now.


I agree with @shanaqui and I do think eventually there will be a way to do this in a non-nightmarish way!

Here’s an except from a relevant internal discussion, the resolution of which is probably a prereq for generalized weekends-off:

You know how hideous it is how we schedule updates to bright red lines every week to add flat spots for weekends? It causes a lot of problems to do it that way but it’s proven hard to improve upon. True Breaks is probably an improvement but still means gradually ending up with a monster graph matrix.

What if it always worked by retro-relaxing?

Namely, you skate the edge till Friday, then on Friday your red line gets retroactively easier so that you’re safe till Monday. Starting Monday it’s your normal rate again. Repeat the next Friday.

This has the property that there’s always just one kink in the red line to deal with weekends-off. Your future bright red line is always at the daily rate for the workweek and as you travel along it you just flatten it somewhat so the line in the past reflects the average rate you’ve maintained including weekends. The past line is 5/7 as steep as the future line.

It might be too complicated and magical but it really nails the problem of monster graph matrices so I wanted to throw it out there to help spec out the possible design space.