There's been a decent bit of discussion on the internet recently regarding the place of GTD in modern workflow, starting with Dave Lee's post on GTD's applicability for creative projects and then morphing into an ongoing discussion on its suitability for the modern tertiary sector workflow at all, given the fact that it was designed about ten years ago^[1]. Since I've recently revamped my GTD system to deal with so-called "creative workflows", I thought now might be an interesting time to codify and publish something on the topic.

Unsurprisingly, my workflow got altered a considerable amount after reading Kourosh Dini's excellent book, Creating Flow with Omnifocus, which you should go out and buy because it's awesome^[2]. The problem, as has been pointed out, with many creative projects is that they're not really divisible into atomic, two-minute tasks. If I need to write a chapter of a book, I need to write a chapter of a book. At some point I'll have to spend several sessions pounding out words, and that really doesn't mesh with the GTD ideal of discrete next-actionable tasks. Plus, one of the most discouraging things I've found with GTD is picking a task, spending half an hour to an hour doing it, then coming back to your computer and being unable to check that off.

One of the suggestions Kourosh lays down for dealing with creative projects is a method he calls "touch the keys":

A piano teacher of mine from years past, who I respect deeply and more so with each day that passes, once taught me that even if I have no time to practice in a day, that I should at least touch the keys that day. While it seems odd to just touch the keys, she was right. There is something absolutely crucial to a daily visit in any craft, if only for a moment, to keeping its fire burning.

So here's what it would look like:

1. You set up a repeating task, "crank widgets for 30 minutes" (or even just "crank some widgets", if you apply Touch the Keys in its fullest form).
2. When you have the time, you crank widgets.
3. Once you're done, you check the task. It will repeat tomorrow, but for now it's gone.
4. Tomorrow, you're reminded to crank widgets again.
5. Once all your widgets are cranked (and this is the most inelegant bit of this whole jobby), you delete the task rather than just marking it done.

This is a pretty good solution - you get to make incremental progress towards a goal, it reminds you constantly, plus you get that reptile-brain satisfaction of checking a box. The one problem I find is that there's no agenda: repeating tasks don't give you a Big Plan to work alongside.

There's a couple of ways to solve this (dumping your plan outline in the notes field springs to mind, for example), but here's my current solution.

For each creative project, you have two sub-projects. One is labelled "Touch the Keys", and is simply a repeating task with the appropriate context. One is labelled "Goals", and has a GTD-esque set of next actions (except these next actions tend to be a bit bigger than your average two-minute task). Your Goals stay out of the way right now - they'll only clutter up your task list, and you don't want to see them. What you want to see is your Touch the Keys repeating task.

When that TtK task pops up, it's time to take stock of the project. Switch over to it (this is where perspectives would be really handy - project-view focussed on the Goals sub-project would rock at this), check what's next on your to-do list…and work on it. When you're done working (you run out of time or attention), you check off any goals you've achieved, but mainly, you switch back to whatever your daily perspective is and you check off that TtK box. This project is now out of your hair for the rest of the day.

I have a template set up for this, that looks something like the following.

If I want to start a new creative project, I just duplicate this project, Find+Replace «Creative Project» to whatever the project's name is, and move it to the proper place in my project hierarchy.

I originally set a due-date on my TtK tasks, so that I'd work on each of my creative projects once a week. All it takes, though, is one big week (like last week for me) where you simply don't have the energy to work on the creative projects, and suddenly that overdue badge pops up in your dock. I'm not good at being creative on a deadline, so I either ignore it (red badge sits there, hiding actual overdue things) or just click "done" despite not having done any work. Both these actions undermine the task system: one degrades the importance of due dates, and one degrades the importance of actually doing a task. So now my creative projects don't have any due dates - just an occasional flag to keep them on my radar.

Future work

Right now, the project's Goals are assigned their proper context, which means for me they pop up whenever I'm doing anything. I'm thinking that they might benefit from either no context, or some context like "Creative Projects". That way, if I have half an hour free and I'm feeling creative, I can check down my list and just do something that requires creativity, but otherwise they keep out of my way.

This is an interesting impact of the whole ubiquitous access to data thing: no matter where I am, as long as I have some form of electronic device I can probably work on one of my creative projects.

But why keep creative projects in GTD?

The whole reason this discussion got started was because GTD is designed for projects that break down into atomic tasks, and creative projects tend not to fit this mould. There are plenty of other ways to handle tasks: it's easy to get fixated on GTD. The problem is, a lot of my work fits the GTD mould well, and creating a different system for creative projects adds another layer of complexity to finding out what I need to do. A lot of my creative projects are hobbies or free-time things. I find myself overwhelmed by everything else that takes my time, and if I don't make time for them, and remind myself that I need to do them, they'll end up falling by the wayside. By keeping my projects in my GTD system, they maintain first-class status. And that means I'll work on them more.

---

1. Which I find an interesting transition, since I believe most of the changes in this time (ubiquitous communication and access to data), while helping with the creative process, don't alter the creative process that hugely from where it would have been in 2001. However, at that point I was a teenager struggling with high school rather than clashing priorities, so I may be very wrong in this case. ↩︎

2. Or alternatively, if I fanboy enough on here it seems I'll eventually reveal everything in the book anyway. ↩︎