"Review Attention" or "Priority"

original post by @Folke on May 5th

I would absolutely love having a color indicator on each task, indicating how closely I must watch and consider this task. It is an absolute life-safer for me.

In the app I currently use, Doit, there is a feature called Priority, that I use for this purpose. This was the main reason I moved from Nirvana, actually. I had figured out while using Nirvana what I need a good way to cope with long lists (Next, Waiting, Someday), and people there had all kinds of tricks using Someday or Nirvana’s Later lists etc etc to limit the length of the Next list. But I discovered that what I need is definitely not to move true Next actions away from the Next list. What I need is a way to see very easily which ones I need to look at right now and which ones I can ignore.

The Priority feature offered in some apps (like Doit and RTM) is just what I needed. My standard setting for new tasks is Medium (blue) - a discreet but clearly visible vertical blue bar on the left. I can increase the level to High (red) or decrease it to Low (turquoise).

I always speed-review all blue and red Next and Waiting tasks every single day. I then deliberately ignore the Low ones, which makes the daily scan so much faster and safer. I can concentrate on the right things. That is important - I won’t overlook things that I should be looking at just because they drown among all the rest. It is more than enough to review the Low ones once a week - otherwise I would not have turned them Low.

By having the Low Next actions still on the Next list, rather than stashed away somewhere else, which many people do, I gain the advantage that whenever I filter for something to do (say Errands) I see on one single screen all such tasks that are truly Next, i.e. possible to consider doing right now. That is important, too - I won’t overlook some possible actions just because they are “hidden away” on some other list.

The clear turquoise color for Low also makes it easy to focus appropriate attention to those during my weekly reviews. (The red and blue ones I see every single day, but the turquoise ones may deserve particular attention during the weekly review.)

Having the red color for High makes it possible to identify those (hopefully few) tasks that I never want to overlook when I visit the Next list, but which I am unwilling to put in Focus yet if I am overloaded. (If I have “flow” and am not overloaded, I usually Focus my red tasks almost immediately, unless there is some obstacle, but if I have fallen behind, been sick or at some conference etc, I may have more red tasks than I can possibly focus on right now.) Very useful.

Now, you may ask, what on earth does this have to do with “priority”? The answer is “nothing but everything”:

My criteria for selecting a level (color) is based entirely on the answer “when do I want to be sure that I see and consider this task again at the very latest?”. If the answer is (for a Next or Waiting task) “tomorrow’s daily scan”, then I select Medium (blue). This is my default, as it were. If the answer is “every single time I even open the Next list”, then I mark it red. If the answer is “next week is good enough for review - because I know that I will stumble upon it sooner if a matching context comes up”, then I mark it turquoise.

Now, it just so happens that the ones I want to see more often are usually more important or urgent or critical or whatever than the rest, in other words “higher priority” in some sense of that infected word.

The beauty of having a colored indicator for this is the fact that this “priority” level is always visible in the corner of my eye (without needing to read) and it is equally visible no matter how I keep my lists grouped, sorted or filtered. I always see the color, subtle but distinct. And I look at the color literally all the time as a time-saving and virtually indispensable guide for my attention.

I strongly recommend you implement this.

1 Like

This one is tough - As it feels a little like it goes against GTD principles to me. The whole concept from David’s books is that you determine what you can do right now - based on the context you are currently in, your energy levels, the time you have available. After you apply those filters to your next list you get a snapshot of all work you could do right now. You should be left with a fairly manageable list of items at that point. Then you select the ones you want to do - I like to put them in my focus list at that point - and start to work. Sometimes I will order them manually in the order I expect to to do them. Why does that not work @Folke for you?

I do admit that the one area where I have always felt DA methodology is weaker is around setting priorities. So I’m willing to listen to alternative ideas. But where possible I’d like the solution to map somehow the core GTD methodology.

Some questions:
Would you want this feature for the next action list to pick what should be on your focus list? Or also on in the focus list?
How many others would find some sort of priority system (other than manual order and filtering) helpful.
What are some possible solutions? For example: How about allowing you to set the color of your tags? Then you could set a red tag for high priority.

Other ideas?

Absolutely correct. Totally agree. This the “micro”-situational task selection that is so fundamental to GTD. Priority often has very little bearing here, although DA includes that, too. But usually, if you are at the hardware store, you’ll buy everything that is on your list, not just the important ones, right now while you are there. And if, due to time pressure or heavy total weight or whatever, you have to prioritize, you do that intuitively. It would not be worth the effort to encode priorities just for those odd occasions.

DA emphasizes priority a lot at the higher horizons, even as low as the 10 k horizon - inactivating projects and activating those that are important in view of your higher goals, and so on. Quite ruthlessly.

But DA is remarkably quiet about priorities at the “1-3 days” level (~5k?). When you start a new fresh day DA recommends (obviously) that you look at your calendar to see what appointments you have. Those will form a framework of contexts and time frames that you will have to adapt to that day. And the remaining time you can often make use of with the “micro”-situational approach we talked about above (e.g. if you have 30 minutes in the Hilton lobby, make the best use of that context, and the time and energy that you happen to have). But is that really all? What if you don’t have any appointments (no contexts and no time frames predetermined)? And what if you do not have any objectively true deadlines dictated by others (and no fake planning deadlines etc either, which you should not have)? Etc. Under such open and uncommitted circumstances, how would you select your “framework tasks”, those that will determine what contexts and energies and time frames you will even have available for “micro”-selection later today?

Lets say you have 100 next actions. One of them is “Get doormat replacement” (the old one is getting a tad frayed). Another next action is “Write and send proposal to XYZ Corp” (promised two days ago to send something very special to them as soon as possible, but no date promised, big potential order, have not even started writing yet). And you have 98 other next actions also vying for your attention. Which of these would make you cry more tears if you forgot about it? What if you forgot about it simply because you did not notice it?

I am talking about priority here in the sense of “review attention” - i.e. how necessary is it for me to at least consider a given task before perhaps deciding to do some other tasks despite all.

DA has never (to my knowledge) expressed any opinions about these things.

I agree 100% with DA about virtually everything he does say on priority. For example, you should not make so-called ABC priorities, i.e. place tasks into phases where you have to finish all the As before you can start on the Bs. I never do that, either.

I assume it would follow the task wherever it goes. That would be the most logical and consistent. It would be most essential on the Next list, of course, but I would prefer to have it visible everywhere (including the All Projects view and Focus, and also for Waiting and Someday tasks, and to be able to preset Scheduled tasks).

Absolutely. Especially if it is a red background blob for the tag (easier to see than small colored text). And I’d choose some duller color for Low. And I’d leave all the normal ones untagged both to save work and have a safe default. That would solve maybe 80% of the problem, and I believe I would probably be quite happy with that. But ideally I would prefer to have these indicators in a totally fixed position (easier to see without “scanning”), and I would also much appreciate some further functionality based on these “review attention priorities”, such as grouping/sorting of the Next, Waiting and Someday lists, which is very convenient, but which would require these “priorities” to be recognized as a “data field” in their own right (compare with RTM and Doit if you’d like to see where I got the inspiration). But yes, I’d be quite satisfied with user-colored tags :slight_smile: And especially if I had a Hide filter to eliminate the Low ones from time to time to see the others even better :wink:


I also consider the Priority feature as very useful and handy tool to select the best items for doing now. Otherwise you can overlook one or several important tasks between many other next actions they can be in right Context, Time and Energy but at the same time they are unimportant and can wait till the more important tasks are done.

As I know David Allen recommends the selection of tasks they should be done now by: Contexts (office, home, errands,…), Time (5 min, 15 min, 30 min,… ), Energy (High, Medium, Low) and PRIORITY (High, Medium, Low). Am I right? I think the Priority feature goes not against the GTD. Maybe, the GTD experts like @Proximo & @Folke will correct me.

From the 4 aspects of the tasks (Context, Time, Energy, Priority) I use only Contexts and Priority for choosing the right actions I want to do now. I think I am one of many users who would like to see the Priority feature in GTDnext more than Time or Energy. However I can understand that for many users also the Time or Energy can be important.

@James, what are your plans for adding the Priority feature including Filtering and Sorting by Priority?



I am not a GTD expert by any means, but I appreciate your compliment.

You are correct in listing the 4 criteria for choosing what to work on. Time, Energy, Context, and Priority.

Priority is part of GTD but David Allen implemented it in a more dynamic way than hard coding your priority by a number or high, medium, and low.

The reason for this is that if you hard code priorities, you can create a trap were you are no longer considering the low priority tasks in your lists because there is an ever growing input of tasks that may be higher in priority. This can cause low priority task to never be worked regardless if you are choosing the correct time, energy, and context.

This is why the priority process happens after you filtered your options by the first 3 criteria. You first filter by how much time you have which hides any tasks that fall beyond that time limit, you then filter by energy which hides task that don’t match the current energy level, and lastly you filter by context to hide tasks that are not relevant to your context. What you end up with is a much smaller list of potential task to consider. At this time, the priority is what you use to consider the smaller list of possible task to work on, but it’s a dynamic decision that removes the hard coding process used by many other systems.

This allows you to only consider priority after you filtered by the first 3 criteria, which gives all the possible tasks you can choose from an equal chance at getting picked. Priorities are ever changing and if you hard code them to your tasks, it may not represent what your priorities are a few days later.

This is why dragging tasks is such an important option for any GTD app. When you are looking at the smaller filtered list, you can easily drag the task to the top which currently represent higher priority tasks.

If your hard code them in, you may not consider looking into each possible tasks details to determine if the priority is correct or if things have shifted since you first selected your priorities. Now you stand a chance of working on the incorrect task simply because it was a high priority last week.

In my line of work, priorities change on a daily basis and this is something David Allen not only recognized, but eliminated from GTD as a practice.

Can a GTD app use hard coded priorities? Sure, there is no law preventing you from using them. I would not recommend it myself and years of experience have proven to me that David Allen was on to something when he decided to keep hard coded priorities out of the GTD methodology.

Just my $.02

1 Like


I am not an expert either, but I think David Allen has caused a lot of confusion by his way of using the words priority and importance for many different things, some of which he strongly recommends and some of which he strongly advises against.

For example, just as @Proximo points out, David advises against using hard-coded priorities for the typical task selection situations that occur during the course of the day. It is simply not worth the effort. Once you have narrowed down your choices that much you usually have no problem picking the most appropriate tasks. You can weigh in priority (importance) easily enough anyway. In fact, David makes no assumptions about hard-coding for energy or time either. The only truly hard-coded factor in GTD is context. This can be hard-coded even on paper by using separate next action lists.

Another thing that David strongly advises against is what you might call “artificial hard sequencing” based on priority (or ABC priorities), whereby you create a pipeline in, say, three stages and “refuse” to touch stage B tasks until you have finished the stage A tasks. As @Proximo points out, such staging is very bad, because it makes you ignore perfectly suitable tasks way too long, maybe even forever. Even the low priority (low importance / low urgency) should be among your options when you make your task selection based on context, time, energy and priority.

David also points out that priorities can vary a lot during the day. He is then referring to priority in the sense of what you eventually select to do first at that time of day in that situation, not to the task’s importance or urgency as such, which is something quite stable. For example, if you are running dangerously late in preparing a proposal to a client, then this is a “fact” that is not influenced by your currently having breakfast or having ten minutes before heading into some meeting. But your choice of tasks to do right there and then will be heavily dependent on such circumstances.

David Allen strongly emphasizes priority/importance, though, but mainly on a higher level. He emphasizes again and again that you must do the important things first. He even advises you to put lower priority projects on hold etc. There should be no doubt that David understands when priority counts, and when it doesn’t.

One thing that David seems to have forgotten to talk about is the relevance of priority when planning your day as a whole. He mentions looking at the calendar to know where you are going to be at what time etc. The calendar actions will form the skeleton onto which you will select and add various next actions that match those situations. But even if you have no calendar actions that day, you still need to select where you are going to go, what kinds of contexts you will set up (e.g. equipment, errands, certain people to get hold of) etc. At this stage, a firm-coded priority (importance/urgency) is very useful. For example, it helps you notice the proposal to that client that you are running dangerously late with. Even if you are stuck in meetings for two days you at least do not have to worry that you will overlook this important/urgent task.

GTD is based on reviewing, even daily. I review my Next and Waiting actions every day (in addition to all the times I see many of these tasks again due to filtering or AoR focusing etc). But actually I do not need to review all these actions every day. Some have such low importance/urgency that it is sufficient to consider them whenever they show up while filtering, or during my weekly review. These I mark as low priority and thereby save a lot of unnecessary daily review time. And even more importantly, it helps me direct my daily review attention to those medium and high priority tasks that may have more need for it.

I dare say that nothing of what I do goes against David Allen’s recommendations.

@Folke has very good points as usual.

The bottom line is that Priority is part of GTD no matter which way you look at it. It’s up to you to decide if the traditional hard coded method works best for you or the dynamic method David normally pushes for. Once you understand the pro’s and con’s of both, you can determine what works best.

I have no issues with a GTD app providing hard coded priority options because I can personally ignore them. Just be careful with adding too much to any one app or service because you can quickly over complicate the app and the user experience can suffer for it.

Pocket Informant and Toodledo are good examples of this. I have Pocket Informant on my iPad and used it for a while. The massive amounts of options and things you can turn on and off is what drove me away from using it. They tried to be a jack of trades and in the process they also became the master of none.

I keep things as simple as possible while making it functional for my needs. This is what everyone must ultimately figure out. We work in different environments and what may be a simple approach for me, may be an impossible approach for another.

If things work out well for me, in the next week or so, I will complete over 172 next actions. :smile:

Thank you very much @Proximo and @Folke for the instructive answers. I do have to agree with most of mentioned.

However I would like to explain better why and how I would like to use the Priority in GTDnext as well as why the Priority feature can be helpful for many other users.

I have used the GTD system for several years and have to say that I do not like the hard coding Priority. My favourite apps were / are NirvanaHQ, Doit.im, Zendone and now I see a big potencial in GTDnext. Maybe my GTD workflow is not as pure, because I try to use several “not GTD” ideas like Eat the frog, MIT - Most important things, some thoughts from Leo Babauta etc. But I really like the ideas of DA and GTD and think that my system is mostly GTD :slight_smile: .

What I am missing f.e. in Nirvana HQ is something what @Folke explained very well and can be better solved f.e. in Doit.im… I also do not want to give the true Next actions somewhere else from Next list to reduce the no. of tasks in Next list if the tasks there are all really the true Next actions. However if I do it this way, I have normally around 200 work Next actions + around 70 personal Next actions.

If I want to choose in my office several times a day a couple of right actions for Focus/doing now, it takes uselessly a lot of time to read again and again the same long list of all next actions. If I know that maybe 60 % or 80% of the Next actions I will not choose for doing now, today, tomorrow, maybe not all this week, I do not want to scan them again and again several times a day, each day. I would like to check such kind of Next actions once or two times a week or during the weekly review.

As @Folke described a good way to solve this problem could be to mark some tasks with color. It does not mean that when I am in the @office, @errands or with some @person that I will check only the “red” tasks with high Priority. The higher “priority” can just help me to check more often, faster and more carefully “the more important” tasks and less often the other tasks.

I do not have the problem that the less important tasks will be never done. Many times I want to work and work on less important tasks, but at the time it is important for me to be aware that there are also some unfinished really important tasks that I should finished today or asap. Some people mark such kind of important tasks with a Due date but I think if the tasks are not really due, it is very unlucky way. I have to agree with @Proximo that the dynamic way of using Priority by manually drag and drop the tasks in order by Priority is more realistic and probably to best way…

Anyway, to read my 200 Next actions several times a day is useless, to use Due dates is not a good way to solve this, to filter by Contexts - @computer, @phone, @office, @online,… cannot help with reducing no. of tasks many times, because the most of my day I am in the office, with my phone, computer and I am online all the day… I can just filter out some tasks that can be done @errands but there will be still 170-180 Next actions that I have to read again and again. To choose by Time and Energy gives for me personally not a big sense, because if there is an important or urgent task which should be done asap, it is better to start doing on it in most cases asap and not tomorrow when my Energy will be better (or not:) and I will have maybe more time before a meeting than today. There was a time when I used the Energy feature in Nirvana HQ as the missing Priority feature - several Tasks/Next actions that were urgent or should be done asap were marked as High energy, “Normal” Next actions - no energy level, Next actions that can be done today or in 1 month and will be probably never important or urgent - low energy. I scanned several times a day the high energy and no energy tasks and once a day or sometimes once a week the low energy tasks.

I do have sincerely agree with @Proximo that the Priority feature can be used by some users in not GTD way if they want. But it can be done with all GTD apps in General as well. However I still think if the Priority would be implemented in GTDnext it can really help the people with many Next actions and it can be used in right GTD way too.

… sorry for my English, I am just beginner therefore I hope you can understand the most of what I sometimes awkwardly wanted to say :slight_smile:

@zdeno I agree with you:

In fact, when I apply a “priority” in Doit I do not even think in terms of priority or importance or urgency or sequence or anything like that. I think strictly in terms of “How often do I want to seriously review this task on a routine basis (in addition to all the times that I might stumble across it while filtering or while focusing on a particular AoR or project etc)?”.

This is a very simple question to answer; I do not need to get bogged down in questions about how to represent some particular sense of importance etc. I just need to decide “How soon do I need to see this again?”.

For Next and Waiting I attach these meanings:

  • Low: Review once a week is more than enough
  • Normal/default: Daily review
  • High: Every single time I open this list, no matter how often, several times a day. I usually have very few tasks here, unless I have been sick or away or overloaded or unproductive or have promised too much to too many.

For Someday/Maybe:

  • Low: Every four to six months or so
  • Normal/default: Weekly review
  • High: Every time I open this list, preferably a few times a week if I have time. These “High Maybes” typically are some super tempting things that I am afraid might be too risky, but which I have not given up thinking about.

The funny thing is that this simple three level color encoding of “review how soon again” corresponds very closely to all practical aspects of importance, urgency and so on that I can think of. If a certain task were neither urgent nor important, why would I want to review it again so soon? So this super-simple decision (how soon again; without having to fiddle with any stupid timers or dates or anything) not only tells me what to look at during this particular visit to the list, but it also gives me a very clear signal at a single glance of what is important and urgent, and what is less so. And I do not have to hide anything away using separate lists or artificial “reminder timers” or “review schedule” etc. Everything is always right there right in front of me, always visible no matter how I group or sort my list. In clear color. Simple and perfect.

@Folke, very good tips. This is exactly the way I would like to use the “Priority” in GTDNext. Maybe the name of this feature should be different from Priority. Maybe the Review would be better. But the name of the feature is not important. More important is if the developers think this feature can be useful and will be implemented.

1 Like

This is one of the best threads on the forum to date. Some really insightful ideas from everyone. Thanks and keep it up!

Good blog post on prioritization, @James: http://blog.gtdnext.com/prioritize-work-using-gtd/

The post very well describes prioritization in one of the two main senses that David Allen uses that tricky word. Perhaps we could refer to this as the “situation level” prioritization (a time frame of no more than than a few hours), when you are in a given context, energy etc and need to decide which action to do first. In such situations it is usually quite easy to look at the possible actions that match that situation and simply pick (intuitively) the actions that seem most important or critical or suitable etc. I agree with DA that there is no need, or even any use, for representing such priorities in any way in the organization or “coding” of your tasks and projects. Simply focusing the relevant tasks, and optionally placing them in a sensible order, is just what you need, just as you describe it.

David Allen also talks about, and emphasizes the importance of, having very clear and firm priorities in the top-down sense, from your higher horizons and down to the project level. This “top level” prioritization (a time frame of at least a week and often much more) requires some degree of organization and/or memorization. For example, DA talks about measures such as selecting a reasonable number of projects to be active (“prioritized”) during the coming week.

In my opinion, there is yet another, intermediate level of priority that DA forgets to mention explicitly, the “day level”. He does mention that you start each day by identifying the calendar actions (appointments) that will form the “framework” for the day (which sets the contexts for everything else that you might decide to do later that day), but in addition to calendar actions I know that for most of us there are often also other actions that are of critical importance or urgency and really should be done very soon indeed, perhaps preferably today. I personally find it useful (indispensable) to be able to “flag” such actions as “high priority”, simply in order to make them easier to find (harder to inadvertently overlook) and it is in this intermediate sense (the “day level” - the “next few days” perspective) that I use a Priority field (and happily refer to it as “review attention” to avoid infected controversies). Conversely, my “low priority” actions are those that I do not need to include in my daily review - a weekly review is enough. This is an important time-saver for me, which also helps reduce the risk of these actions obstructing my view of the more important tasks. My “normal priority” ones I scan/review at least once per day, as per normal GTD.

1 Like