David Allen says "The "ABC" priority codes don't work" but he doesn't explain exactly what system he means by "The "ABC" priority codes" or what his criteria are for it not working.
I want to make two points: first, that recording priorities on task lists can be useful and worthwhile, without necessarily making specific commitments about how to use the information. Doing this is quite compatible with Allen's "Getting Things Done" (GTD) system. Secondly, it sometimes makes sense to step away from the flexibility of GTD and make a rule or promise to yourself about doing the higher-priority ones first. This can be worthwhile as a motivating technique even if it sometimes has a cost in efficiency.
I think GTD is fantastic. I use it (more or less), I've benefitted greatly from it and am grateful to Allen for having devised it. One of the things I like is his advice to sift tasks first by context, then by time, then energy, then finally by priority -- a method which may seem counter-intuitive at first, but actually makes a lot of sense and works very well in practice. However, Allen's system is not the only useful one.
I don't disagree with any of the arguments Allen presents in the newsletter post; I just don't see how they add up to not doing ABC prioritization. Perhaps what he's really arguing against is a system in which you aren't allowed to change the recorded priorities.
Recording priorities takes time, but can save time later because you don't have to read your whole task list every time but only the higher-priority items, and you don't have to keep recalculating priorities when there's no reason for them to have changed. Depending on how the information is used it might sometimes lead to sub-optimal choices, but can also lead to better choices.
Priorities you've previously recorded are sometimes better than spur-of-the-moment thinking. Which is the more accurate judgement: the way you feel when you're considering starting a difficult task, or the way you feel when you're somewhere else, suffering from the fact that this task hasn't been done yet? Although your judgement can improve with time as situations change and you collect information, it can also waver from optimal when you get too tired to think clearly or when the effort of doing something now seems to loom larger than benefits distant in time and space.
I agree that if you're tired after some meetings, that may be a good time to water your plants; however, I don't think you need to re-evaluate your priorities to come to that conclusion. That comes from an earlier step in GTD -- the sifting by energy available.
If priorities change with mood, context and time of day, then maybe the original priorities often repeat again, Then a priority marked on a task list can be taken as a polite request from yourself to yourself to please show some consideration for the part of yourself which made the priority mark and which will also later experience the effects of the action having been done or not done. It can be gratifying to experience things going comfortably, easily and efficiently because you've done things for yourself.
When I mark a priority on a list of actions, it doesn't mean "do this or else!" The usual sifting by context, time and energy still occurs. It usually just means that that's the priority level of that action -- or at least was an estimate of the priority level at the time it was written. David Allen says in his books that you can re-negotiate agreements with yourself at any time.
If you're going to change them later, why write them in the first place? First of all, to save time so you don't have to read your whole long list of actions every time. Secondly, because it cues you to take a good look at whether you're changing the priority for a good reason or not.
Allen's argument about watering plants makes more sense to me if what he's arguing against is rules about doing higher-priority tasks first. While the flexibility of GTD is usually an asset, sometimes it helps to be able to commit yourself to following a rule.
If you've promised yourself that no matter how frazzled you are after all those meetings, you're going to write at least a draft response to that difficult but important email before doing anything else, then rather than watering your plants it may be best to either write the draft email or, if you're really too tired, to just rest and do nothing, in order to maintain the integrity of your system, your confidence in your own follow-through on promises to yourself, and your determination to write that email within the next couple of days before your plants wilt. Allen may consider this a waste of time, but for some people, these sorts of arbitrary rules can be worthwhile as a way of getting motivated to get the most important things done; for example, for some of the people who posted comments at the bottom of this page.
Anyway, GTD already has a three-level priority system! "A" is work you do as it comes up. (If you burn your hand, you stick it under the cold water without writing that action on a list first.) "B" is your "next actions" lists, and "C" is "someday/maybe". GTD also includes the idea of making commitments to yourself: I'm pretty sure one of the books includes the idea of making an appointment with yourself -- that looks like a commitment. He also says you can re-negotiate your commitments with yourself at any time, which kindof implies you have commitments to yourself.
People are free to use ABC priority codes if they find it works well for them.
The following is a slightly exaggerated comparison of how I would handle a task as compared to David Allen's GTD as I understand it. See also Sorting actions by energy level required, etc. for a description of my system.
Suppose I'm riding a bus and think of something I want to do at home sometime in the next few weeks. I pull out my notebook and immediately write it on my home actions list, in the section for low-priority actions that take significant time and energy. When I'm at home and have time to do stuff, I look at my list, often reading only the high-priority or (if I'm tired or short of time) only the easy/quick actions. I may read that action a small number of times over the next few weeks, and eventually do it.
Total effort: Writing once; reading a few times; doing once. Three steps, one of which is repeated a small number of times.
The initial writing involves mental effort to assess the level of priority and difficulty.
In David Allen's system as I understand it, it would work like this. I'm on the bus and think of the item. I write it on a slip of paper. When I get home or to the office, I put it into my inbox. Later, I process it and decide it can be handled with a single action (no project needed). I add it to "Someday/Maybe" because it doesn't have to be done this week. Then each week when I do weekly review, I look over it and decide whether to put it on my list of actions for that week. One week I predict that I'll likely have time, and put it on my actions list. Oops, it turns out I don't have time. On the next weekly review I leave it on my home actions list again. For those two weeks, whenever I'm doing stuff at home, I read all the actions on the list, select the ones I have time and energy for, (usually rejecting that action at that stage), sort them by priority (sometimes sorting that action as low-priority), then do. Eventually that one gets done.
Total effort: Writing once, moving into inbox, processing, reviewing several times on Someday/Maybe, moving to actions list, reviewing a large number of times on actions list which involves deciding on time, energy and priority levels, doing. Seven steps, two of which are repeated.
Mental effort is required when processing, when deciding to move off Someday/Maybe, and when reviewing on the actions list and deciding about time, energy etc.: three steps requiring significant mental effort as opposed to one the way I do it.
Actually, in Allen's GTD it wouldn't work that way, because only projects go onto Someday/Maybe. I don't know how he would handle a single action that can be done sometime in the next few weeks (or months). Maybe he'd actually do something more similar to what I do, but that isn't explained.
My method takes more mental effort when initially writing it down, but I find that it usually only takes a few seconds and is doable. A lot of mental effort is saved by putting it immediately onto the right actions list and by not having to think much about time, energy and priority except when initially writing it down.