- In search of the ideal task management solution - part I
- In search of the ideal task management solution - part II
You can consider this part III.
I have been using the Todoist approach with The Cycle outlined in Tom Limoncelli’s Time Management for System Administrators since about December of last year.
I originally planned for this post to be a monolith outlining all my rules, but my rules haven’t really changed or gotten more refined since the previous post. Instead, I’ll outline what has worked and what hasn’t worked. I have some lessons learned from this process, and still plenty more to learn.
This has been working a lot better than previous attempts at task management for me, but that’s not to say it’s the be all end all.
Current approach
I wanted to have a clean set of “rules” that I operate by, but I think it’s more logical to arrange them in a typical flow for a workday. (Weekends are largely the same, except I shouldn’t see any work tasks.)
- Take 10-15 minutes at the beginning of the day to plan the day. Anything that MUST be done at a certain time takes priority. Anything that MUST be done today, but not at a specific time, takes second priority. The rest of the day’s time is spent 1) fighting fires from tasks that come up or 2) doing less priority tasks.
- Schedule tasks that must be done at a certain time and add reminders if necessary.
- Prioritize tasks that MUST be done today, but not at a specific time.
- Schedule any personal tasks to a time after the workday, so they do not appear in the “unscheduled” view alongside work tasks.
- Throughout the day, add new tasks to the Inbox. I
- I should not worry about anything except priority when adding a new task, especially if I am working on something else. Generally, I only make it priority 1 if it’s something that takes precedence, otherwise I leave the priority as default and figure it out later.
- Intermittently, if time allows, review the Inbox - sometimes I’ll need to add tasks to the Today view. But the most important thing is that the task is now in Todoist and not in my brain.
- Take 10-15 minutes at the end of the workday to manage ALL work tasks.
- Mark anything off that is completed or no longer relevant.
- Reschedule anything that did not get done to an appropriate future date.
- Clear any work tasks from the Inbox - add a priority, due date, and move it to the Work project.
- Proceed to the remainder of the day. Nothing changes here, same approach applies.
- Take 10-15 minutes before going to bed to do “Inbox Zero” for personal tasks.
What has worked
Utilizing core views of Todoist instead of projects
I wrote about this being one of the core changes to my workflow. It’s been an absolute life changer. While organizing tasks by category or project can be useful in several ways - reviewing progress on a particular project, being able to see all tasks related to a given project, etc - it’s not worth the administrative overhead to me. It’s very easy to determine (with few exceptions) whether a task is work or personal, and it only goes into one of those projects after it has a due date, so I know I’ll never lose it.
There is an important exception - when I’m working on a particularly large project, I like to have a separate workspace. I tend not to add any due dates when doing this, and just leave a generic task in my Today view (e.g. “work on project X”) and organize the project itself by other things, such as custom ordering, priority, and project sections.
Another exception would be oddball lists such as habit trackers, shopping list, etc. I try to keep these pruned, and still follow the same Inbox approach (chuck the task into the inbox, organize it later) so I’m not sitting there trying to remember what project I need to put this task into.
Deadline feature
Todoist added this feature fairly soon after I switched to this approach. At first it was a cool way to add a bit of extra metadata to the task, but it really came in handy when my tasks started to overflow.
If I have 50 priority 1 tasks - meaning all tasks that need to be done today, if possible - how do I even begin to prioritize? Deadlines. If 10 of those need to be done immediately, add a deadline of “today” to those tasks. They filter up to the top of the priority 1 tasks. If 10 more need to be done by the end of the week, add a deadline of “Friday” to them.
This tends to make priority less important (pun intended) as metadata about the task, but my tasks can get so out of hand that it’s more important to spend time knocking out tasks than asking myself “does this really need to be done TODAY?” and futzing with the priority. Just leave them be, but take tasks that REALLY need to be done and filter them to the top.
This ties into the first struggle that I’ve had.
What has not worked / current struggles
Too many tasks
About a month ago I started drowning in tasks. Not only did I have 75+ tasks on any given day, but too many of them were very important, and I had been overutilizing priority 2 so I had 30+ tasks in that category. (I mean, they were all priority 2 - need to be done “soon”, but certainly some of them should have been done sooner than others.) I would open Todoist, get overwhelmed, and freeze up. Either wasted a lot of time either sitting there not doing anything and trying to pick a task, or fiddled with the tasks trying to order or prioritize (and inevitably getting interrupted or distracted).
After a couple days of gasping for air, I whipped out the good ol pen and paper. At first, they were completely incompatible. I opted to use the paper instead of Todoist. I paid for it at the end of the day, my lists were getting out of sync, but I was able to manage through the workday, which was more important short term.
As this workflow progressed, I realized that I could marry the two. Still spend time organizing tasks in Todoist, but don’t fret. Utilize deadlines as needed at the beginning and end of the day, but then pick a task, write it down on paper, close Todoist, and do the damn thing. Cross it off, open Todoist, pick the next task, and do it again. And never write more than one task down at the same time. If I get interrupted with something else to do, add it to the inbox. If needed, I can do that next.
My struggle with dedicated paper lists is if I lose it or my cat pisses on it, I’m screwed. Not to mention trying to find a task from a month ago, or trying to schedule a task for myself a month from now. In this workflow, I treat paper as disposable. It’s a tool to help me focus and give me some physical satisfaction when I complete a task. Todoist is the main organization tool, but I should only visit it when I am picking a task, or in the brainspace to manage tasks at the beginning or end of the day. Bonus - I get to shred the paper at the end of each day. Man that feels good.
Other struggles
The foregoing was the most obvious and painful. The others are less so.
Setting goals
The lines can be blurry for goals. I manage with an approach of tasks and subtasks, but oftentimes I do a task that I stuck in my “Personal” project that should be related to a particular goal. I run into the categorical organization problem. What goal is this related to? Not only do I have to answer that question, which isn’t necessarily straightforward (e.g.: this should be a goal but is not, maybe this goal should be narrowed or broadened, this falls under multiple goals, etc), but then I have to move the task to the Goals project, stick it under a particular goal, set a due date and priority…
I think my approach here should be to stop sweating it. Do my goal reviews as normal, sprinkle in some tasks to happen on holidays or weekends, and anything else that comes up just put it in Personal and do it.
I may get rid of the Work Goals project entirely. I never have time to manage it, and I end up with task drift. Plenty of goals that I set are already completed, or forgotten about. I’m not sure. I never use that project at this point, and I don’t know if that is an issue.
When to refactor or remove tasks
Tying into getting overwhelmed.
A recent actual example - maybe I have 10 different things to discuss with my boss, that came up over the course of a month, all at different priorities. It’d be much easier to group those tasks as subtasks and do them all at once.
My solution this time was to move all of them to Inbox, add a parent task, set it to priority 1 and date it next week, and move all those tasks as subtasks (removing their due date). Now another problem could be I can’t cross that parent task off until I’ve discussed all of the above with the boss, and some of them are lower priority. This time, I made two different parent tasks, set one as a higher priority and the other as a lower priority.
I’ve also resorted to aggregating it onto a sticky note before I go talk to him.
Another thing that’s come up is crossing off a large task with many subtasks after adding the remaining subtasks as separate tasks - “finish up - something or other”. That gives an indicator of my progress on that task, I can cross the big chungus task off, and schedule the remainder accordingly.
But my overall point being: it’s not clear to me when and how I should refactor. Still working on approaches to this. Maybe the solution is just what I’m already doing - intermittent refactoring with an approach that suits the particular situation.
Calendar projects
I initially defined the calendar as tasks with a due date and time and duration, such as meeting or obligation that must occur at a set time.
My question is do I even need that separation? I’ve barely used the calendar filter/view.
Some tasks straddle the line - I’d like them to pop up in my calendar filter, but they don’t necessarily fit that definition, like birthdays. Maybe I’d like something to show up there but I added it to my work project instead of the work calendar project. I am unsure if it’s worth having a different project.
Todoist UX issues
For reference, all these complaints are with the windows client.
Todoist having this level of functionality has been great, don’t get me wrong, but the Today view has been under construction for a seeming long time. Default duration changed to 30 minutes instead of 15 about a month ago, meaning a lot more awkward clicking and dragging when planning my day - then fixed itself, and I’m not sure if that was intentional or just an oversight or some weird part of my workflow.
Other annoyances include:
- Tasks zipping down to late in the evening or up to early in the morning when that is most certainly not where I dragged that task.
- The today view scoots up and down when hovering. It’s hard to describe what exactly is happening, but I can’t get this to happen consistently. Sometimes I will misclick because the task view just moves out from under my cursor. Seems like it would be an easy fix, but I’ve been dealing with this for the better part of a month after a recent overhaul.
I imagine these will get fixed, but it’s also frustrating to deal with them. And what might get changed that I depend on or take for granted as part of my workflow? Oh the woes of using proprietary software for mission-critical.
Wrap-up
As much griping as I feel I’ve done in this post, overall I’m satisfied with the progress I’ve made with this approach.
And for cloud SaaS, I have remarkably few complaints about Todoist. (The majority of the bitching here is more about methodology than it is about the technology.) My ultimate goal is still to replace it, but I’m beginning to think that may require more of a lifestyle and mental shift as opposed to some software silver bullet. Sometimes I feel exceptionally needy when I outline my software requirements.
EOF