To Think About . . .

The price of inaction is far greater than the cost of making a mistake. Meister Eckhart

 

 

 

My Latest Book

Product Details

Also available on Amazon.com, Amazon.fr, and other Amazons and bookshops worldwide! 

Search This Site
Log-in
Latest Comments
My Other Books

Product Details

Product Details

Product Details

The Pathway to Awesomeness

Click to order other recommended books.

Find Us on Facebook Badge

Discussion Forum > High Speed Time Management

I previously posted on the forum about my problem of working too slowly.

This is a follow-up. I’ll try not to make this post redundant; apologies if that’s the case.

(My title is a nod to an old blog series of Mark's: "High Intensity Time Management.")

Self-diagnosis is hard, but I’d say speed of work continues to be my main time management problem, along with related issues such as (a minor case of) perfectionism and occasional procrastination. Right now, unfortunately, a lot of work has piled up and I have not yet found the inertia to get it out of the way.

Looking back on some of my more satisfying work experiences in past years, I realize that I derived a lot of satisfaction from doing work FAST. It was as if in those moments I had made it my mantra to do as much work as possible in the shortest amount of time. It was an exhilarating feeling to achieve something like that… like the feeling some people get while driving fast in a car. It was satisfying to finish work quickly, and satisfying to see how pleased my boss was at my efficiency.

(To answer the obvious question: “Why can’t you achieve that kind of efficiency now?” In part, it’s because my former job was much more routine than my current work. Back then, I could ramp up the speed more easily.)

Now, I’m not saying that speed is everything in time management (and obviously it isn’t everything in life). Mark’s own conception of time management is much broader than mere efficiency, a point which is clear from his writings and which he underscored in the recent thread: “What Would You Do with the Perfect Time Management System?” But speed is something to bear in mind (along with the quality of your work and the results you achieve). And for some of us, a lack of speed gets in the way of loftier aims and more important things. When this happens, speed becomes critical whether we like to admit it or not.

Needless to say, any of Mark’s systems will help with speed. Most of us, myself included, have experienced huge benefits first hand. But I’m tempted to say (though it pains me) that Mark’s systems may not provide a quick and total fix to the issue. For instance, it’s possible to work rather slowly in Simple Scanning. You might work on the same task inefficiently for several sessions, reentering it at the end of the list. And you might get distracted by more fun/unnecessary/trivial tasks on your list. (When I talk about getting as much done in as little time, I mostly mean “as much WORK.”) The system, worked properly, will probably steer you away from such behavior—and, to be sure, Simple Scanning is a great system. But it’s still possible for a user to muck it up, just as it’s possible to drive 5 kms/hour down the highway in a Lamborghini.

So, an update on my progress… I’ve been using Dreams the past few months. It has yielded new insights and clarity about my goals. I have made significant progress on several matters in my personal life (the kind of stuff that can be hard to improve using a long list alone). Dreams has been a unique book for me. I’ve found its low-pressure/relaxed/positive approach congenial. All great things… I admit, though, that it isn't always the easiest system to implement. And I still struggle with the more mundane issue of getting work done quickly.

In a previous thread, Mark recommended timeboxing, which seemed promising, since it was a gamechanger for him in his first book, Get Everything Done. To be honest, though, I haven’t drawn to timeboxing. I don’t like running timers constantly. And more importantly, I haven’t seen huge gains. Using a timer has seemed to me somewhat like imposing an artificial deadline. The brain knows there isn’t a real limit involved. Another timer could always follow, on to infinity. If the mind feels this way, the timer can become a way to mark time only, not an efficiency booster. I stress that this is only my personal experience and I might be doing things wrong.

In searching Mark’s writings on speed, I came across the following quote from 2017 (which I will pull out of its original context): << One of my contentions has been that it’s the psychological attitude that counts as much as the system itself. >> http://markforster.squarespace.com/blog/2017/7/13/high-volume-high-speed-low-resistance-5.html .

This might be true, at least as far as the long-list systems are concerned (Mark was talking about Simple Scanning). Perhaps one needs to have speed as a goal and not simply to “fall back on” a long list. I don’t know. Very recently, Mark has said “The sequence is System > Ambitions, rather than Ambitions > System.” Perhaps the logic of this last quote is preferable?

So, what am I going to try next? I’m going to try two simple solutions. Solution One is to stick to Dreams but to make my future vision (at least for the immediate future) mostly about speed. Really concentrate on that, focus on the positives, and try to achieve pull mode.

Solution Two is to give “What am I resisting not doing?” another try. I used this as a stand-alone question with no list last Lent Challenge. It kicked my butt, occasionally stressed me out, and was fairly mentally taxing (asking myself the question repeatedly). And yet, I really got a lot done. Basically, an unparalleled amount, polishing off a long report early like I never do. So, it may be worth a try again!

(A third solution would be to hire someone to be my "boss," closely monitoring my progress on everything I do. Just kidding... well, half-kidding. I think this is part of my problem.)

We’ll see what happens, but I hope to make progress in the near future.

This post is more self-centered than I would like (I have no desire to describe my life on the Internet). I write mostly for selfish reasons—to see if anyone has suggestions for me to achieve rapid speed (again). But I also hope this might help others with similar frustrations. And there might be something of minor interest here for thinking about time management in general.

On this last point… For clarity’s sake: I’m not saying that speed is all, and obviously, much of value in life comes from taking things slowly. All I’m suggesting is that speed has its place—if only to get work done so that we can move on to those slower moments.
February 6, 2022 at 20:07 | Registered CommenterBelacqua
P.S. I might have added a sentence to the above to emphasize FINISHING things (quickly), not merely starting things. That might be an important distinction to keep in mind.
February 6, 2022 at 20:39 | Registered CommenterBelacqua
Just to make sure I understand you clearly, what exactly do you mean by "speed"?

Do you mean "finishing each task that you start, as quickly as possible" or something like that?

For me, two things help with this.

1-- Look at overall system flow, rather than speed of any individual work item in isolation. Flow of work (choosing the highest value things and completing them as quickly as possible) and flow of energy/attention (maintaining high momentum and capacity, and applying it to the right things).

2-- Focus on finishing. This keeps WIP low, which helps you focus on the things already started, which helps get them completed faster.

I personally have more success with systems that have a bias toward finishing things rather than a bias toward starting things.
February 7, 2022 at 3:44 | Registered CommenterSeraphim
Seraphim:

<< What exactly do you mean by "speed"?

Do you mean "finishing each task that you start, as quickly as possible" or something like that? >>

Yeah, I'd say that's spot-on. "Finishing" does seem important in this context.

Thank you for the suggestions.
February 7, 2022 at 4:09 | Registered CommenterBelacqua
I think there is a lot more to speed than has so far been discussed here. To increase the speed of getting one's work (in the widest sense) done, there are the following factors to consider:

1, The rate at which the work on an individual task is done.
2. The length of the gap between stopping one task and starting the next.
3. The consistency of 1 and 2 throughout the working day.
4. The efficiency of one's working practices, e.g. that one is not doing work in a disorganised or unnecessarily complicated way.

No. 1 is independent of what time management system is being used, and is largely a matter of practice - just like practising running or playing the piano. This may well be one of the most neglected parts of time management.

No. 2 is crucial. The time spent deciding what to do next and transitioning to it can be considerable. All the effect of No, 1 may be lost if there isn't an efficient transition between tasks. Note that time for rest has to be included, but this is best treated as a task in itself, rather than as part of the "gap".

No. 3. The biggest enemy of efficient time management is lack of focus and drift. This can happen either with No. 1 or No. 2. They effect each other. If one's rate of work on tasks slows down, the transition between tasks will slow down as well.

No. 4. This reinforces the point I have often made that good routines are at the foundation of good time management. You can have everything else right, but if you are doing wasteful stuff in a wasteful way, you are still wasting your time.

All of these apply whatever time management system you use. No doubt some systems suit some people more than others, but your speed of doing your work depends more on these four factors than your choice of system.

However there is one exception to this. Changing systems all the time is the one certain way of ensuring that you never get 1-4 working properly. Speed comes out of consistent practice.
February 7, 2022 at 12:49 | Registered CommenterMark Forster
Thanks, Mark. This really gives me a lot to think about.

I suppose I was mostly thinking in terms of what you call Factor 1: "The rate at which the work on an individual task is done." I didn't consider the other factors so much, but now that you've identified them, I can see that they would be important too.

I'm not quite sure how to respond to two things you've said:

<< No. 1 is independent of what time management system is being used, and is largely a matter of practice - just like practising running or playing the piano. >>

<< your speed of doing your work depends more on these four factors than your choice of system >>

On the one hand, these quotes seem totally right to me. Above, I gave the example that it's possible to be slow while using Simple Scanning (a system which, in my view, enables a very fast pace of work).

On the other hand, I'm not sure... It seems to me that your systems do much more than just manage tasks or help you decide what to do. Among other things they (tend to) increase the pace of work. I wonder if you're selling your systems short here. Perhaps Factor 1 (or the others) is not entirely system-independent.

An example from my post above was my use of "What am I resisting not doing?". It seems pretty clear that the question greatly boosted the rate at which I finished tasks, and this increase was rather sudden. It wasn't as if I "practiced" (like practicing the piano) or improved in any notable way outside the framework of the system.

Other examples could be drawn up. E.g., one might encounter a speed increase using Superfocus because the system gamifies finishing stuff and one needs to work quickly to avoid Column 2 clogging and the whole system falling apart.

If Factor 1 (or the others) is (somewhat) system-dependent, then it raises an interesting design issue for you in making TM systems... Will this system boost speed? If so, how? If speed is system-dependent, it also gives hope to people like me who would always welcome an "easy" solution to what can otherwise be a multilayered and seemingly intractable problem.

I may be wrong with the above. I'm just thinking out loud.
February 7, 2022 at 17:40 | Registered CommenterBelacqua
P.S.: I guess a system can't force you to work quickly, and it seems difficult to design one that would generally boost the user's speed of work. But perhaps not impossible.
February 7, 2022 at 18:10 | Registered CommenterBelacqua
Focusing on No. 1, the speed of doing a task itself, I don't have any quick fixes, but it has been my observation over time that speed in this area is greatly trained and increased the more you learn to mitigate any forms of multi-tasking. Multi-tasking or distraction usually comes in these varieties based on my observation:

* Trying to do two disparate things at the same time (resulting in rapid context switching costs and loss of active working memory power)
* Stress from impatience or a desire to have a task done
* Stress or worry about what you have to do or are not currently doing
* Interruptions to your workflow
* Stress from a need to receive novel inputs (compulsions to check email, social media, &c.)
* Attention residue from prior tasks due to improperly eliminating or reducing that residue before starting a new task

When it comes to impatience about having a task done, some might point to the deadline effect as contradicting this. My argument would be that overtime, my experience is that the deadline effect results in the appearance of a temporary short-term gain of output at the expense of long-term speed. I also find the work to be generally done at a lower level of quality, which again tends to cause problems in the future with more work, and thus less total effective speed.

I think the best fundamental thing you can do to train this is to practice your ability to intensely focus your attention for a prolonged period of time on a single idea/thought. This is a whole class of meditations. However, I think this matters across the board. For menial work, your ability to retain focus on the menial task at a heightened level of awareness improves your ability to do that task to a high level of proficiency for a longer period of time. For repetitive automatic tasks, your ability to master those repetitive tasks is directly associated with your ability to isolate and refine your practice to just the specific task you are focusing on (this is where the benefits of excrutiatingly slow practice of things like a tennis swing or golf swing can make huge differences). The more excess inputs are in the system while training that practice, the less sharp the training memory in the brain. For knowledge work, the more you focus on a single task the better you are able to enter a flow state and the clearer your mind becomes, freeing up significantly higher levels of working memory and actually potentially increasing your effective IQ by a number of points, which is directly correlated with speed of solving complete knowledge tasks.

Additionally, your ability to tolerate the isolation and discomfort of monomania of that sort also improves your practical conscientiousness, which is correlated with the relative power to "get things done".
February 7, 2022 at 18:59 | Registered CommenterAaron Hsu
<<speed in this (No 1) area is greatly trained and increased the more you learn to mitigate any forms of multi-tasking. Multi-tasking or distraction usually comes in these varieties based on my observation:>>

I think this list is spot on. These causes of unfocused work are big players in driving inefficiency.

<<When it comes to impatience about having a task done, some might point to the deadline effect as contradicting this. My argument would be that overtime, my experience is that the deadline effect results in the appearance of a temporary short-term gain of output at the expense of long-term speed. I also find the work to be generally done at a lower level of quality, which again tends to cause problems in the future with more work, and thus less total effective speed.>>

I expect this is extremely personal and also depends on the task. Personally I can be extremely inefficient over open-ended timeframes, and setting a concise time is a broadly useful remedy that doesn't necessarily imply sloppy work.

As an idea, your suggestion to train your focus makes sense. I have no idea how effectively I could do that; it is definitely a weakness of mine.
February 7, 2022 at 19:58 | Registered CommenterAlan Baljeu
A rather random addendum to my original post... I described my lack of success with timers, but it just occurred to me that I never properly tested rotating between tasks after timed bursts (as in the original Get Everything Done system). When I used timers, this always entailed a succession of bursts while working on a single task "for as long as I feel like it" (as in the typical "standing out" approach). As reported, the "end-effect" seemed to be lost after a while. But maybe if one is forced to rotate among tasks, this effect is more pronounced (e.g., you'll want to make your 15-minute burst count working on Task A because now you have to do Tasks B, C, and D before you can come back to A).

Any costs of task switching would need to be considered. Perhaps timers would work well with a system like Mark's 3T, 5T, or something like those, so that the number of WIP items stays low... Anyway, this is one more option I can add to my list of things to try.
February 7, 2022 at 20:16 | Registered CommenterBelacqua
Alan Baljeu:

<< As an idea, your suggestion to train your focus makes sense. I have no idea how effectively I could do that; it is definitely a weakness of mine.>>

I think one of the easiest and most effective ways of training this would be with a stopwatch. For example, time how long it takes to answer your email. You'll soon get a feel for what is a good time for the amount of email you have.

Or time how long it takes to finish a set number of tasks. As well as helping you to train your focus on the tasks themselves, it will also show up any inefficiencies in the process of moving from one task to another.
February 7, 2022 at 20:26 | Registered CommenterMark Forster
Alan:

<<Personally I can be extremely inefficient over open-ended timeframes, and setting a concise time is a broadly useful remedy that doesn't necessarily imply sloppy work.>>

I have a pet theory that the reason many people find open-ended timeframes difficult and time limits helpful is that when you have a time limit, you are forcing yourself to suddenly confront the question of what "good enough" looks like. Or, put another way, it forces the mind to clarify what it would mean to complete or finish a task. If someone defines a task without sufficient clarity, they may know what they are working on, but not how to stop. This can put them into an endless cycle of exploration and refinement without sufficient direction. In an open-ended space, they'll never have any external pressure to stop doing this and ask how to refine the task to a stopping point.

In other words, the existence of the time limit forces a fundamental shift in how they engage with a task, but a time limit isn't the actual behavior that causes the task to be finished. Instead, if someone defines a task and clarifies it so that acting on that task is fundamentally moving towards an articulable done state that is concrete in the sense that you can accurately evaluate when you are getting closer to that state and when you have reached that state, then you can work on the task, I believe, just as efficiently, or more so, without the stress induced by a deadline.

The main "Achilles heel" of the deadline, in my experience, is that it tends to discourage thinking and problem-solving optimization, in favor of reliance on existing knowledge. In other words, people will have a tendency to power through something with sub-optimal solutions that they are confident will predictably get them to something, rather than taking the time to speculative evaluate other solutions that will deliver a higher quality result but which they don't know or haven't thought it through yet. The problem is that this free-thinking can go off the rails without the appropriate target, so you must first ensure that you are free thinking towards a done state, and not just thinking of how to "improve" things.

I *do* find time limits useful to limit overwork, but I definitely think they aren't the best way to create completion motivation.
February 7, 2022 at 22:47 | Registered CommenterAaron Hsu
<< if someone defines a task and clarifies it so that acting on that task is fundamentally moving towards an articulable done state that is concrete in the sense that you can accurately evaluate when you are getting closer to that state and when you have reached that state, then you can work on the task, I believe, just as efficiently, or more so, without the stress induced by a deadline.>>

I am a fan of this theory.

<<The main "Achilles heel" of the deadline, in my experience, is that it tends to discourage thinking and problem-solving optimization, in favor of reliance on existing knowledge.>>

This also seems a worthy theory to me. I shall ponder testing these.
February 8, 2022 at 0:21 | Registered CommenterAlan Baljeu
I thought I would post an alternative to timeboxing that encourages completion, which is simply the various forms of closed lists or WIP limiters. If you say that you can't add anything to a list unless you take something off the list first, and you cannot take something off the list unless it is finished, that means that you'll naturally be inclined to finish things on the list to allow for new stuff to show up, if that new stuff has any urgency to you.

I find that this also has the nice side-effect of encouraging you to think ahead about your task and to define the task in a way that is actionable and small (enough so that you can finish the task in the time that you want or close to it). Having an open-ended task on a limited list means that you've just potentially cut off an entire slot on your list for eternity or indefinitely, which at least for me, really makes me re-evaluate my tasks and their size/resolution.

Obviously, this works best when you actually choose a size of the list that is small enough to actually be a limit. The DIT closed list is a form of this, but with some extra requirements.
February 10, 2022 at 1:25 | Registered CommenterAaron Hsu
Thanks for that, Aaron. Use of closed lists seems sensible in this context and that's another thing I can try. Above, Seraphim had also suggested to focus on finishing and limit WIP.

I suppose, in addition to DIT, systems like 3T and 5T (from Secrets of Productive People) come to mind... I have used, for a longer period of time, Mark's "Simplest Form of No-List" (writing down the next thing before doing it). That may be the ultimate WIP limiter. Perhaps the trick with that one is to find the right level of granularity for a task (not too big, not too small)--as we discussed on another thread.

I wanted to thank everyone for sharing your insights on this thread. I'm not sure if there is an easy solution to these issues. Some of the solutions proposed in this thread aren't easy at all (practice until you get faster, improve your concentration, etc.). That doesn't mean they aren't valid. I still wish I could work faster with greater ease. Such as with the ease of switching to a different TM system! And without feeling that I need to do therapy or develop an Olympic training regime. We all can wish!

I suspect there are many other aspects of working too slowly and related issues (e.g., procrastination) which could be raised for discussion. For instance, I wonder if a lot of the problem has to do with overcoming resistance and/or shoring up the willpower to make significant inroads (not just a token amount of progress) on tasks (especially difficult tasks). (Perhaps this was why I experienced some success with "What am I resisting not doing?") Though, like Mark in "Dreams," I'm not particularly keen on the idea of "forcing" yourself to do something (in typical "push-mode" fashion).

Anyway, the time management journey continues for all of us. No doubt we'll have more approaches to try (and re-try), and more things to discuss in the future...
February 10, 2022 at 15:30 | Registered CommenterBelacqua
Belacqua:
<< I still wish I could work faster with greater ease. >>

What objective is blocked by your perceived lack of speed?

Maybe I am off track here, but often things like "lack of money", "lack of time", "lack of speed" are *symptoms* rather than causes. You might be able to find tips & tricks to make a marginal impact on the lack of X, but it doesn't fix the root cause.

The root cause is usually some kind of priority conflict, e.g.:
-- we are torn between A and B, and can't give enough time, money, and attention to either of them, so we feel like we "don't have enough X" -- but the solution is really breaking the conflict between A and B
-- priority conflicts can also lead to overwhelm, which leads to task switching, indecision, etc., which can all generate a sense of making no progress -- "I'm not going fast enough" -- when the real root cause of the problem is the priority conflict

I would tend to go in this direction, rather than just trying to improve raw speed. Based on what you are describing, it doesn't sound like you are lacking in basic skills or raw processing power.
February 11, 2022 at 4:32 | Registered CommenterSeraphim
Aaron Hsu wrote:
<< I thought I would post an alternative to timeboxing that encourages completion, which is simply the various forms of closed lists or WIP limiters. >>

I think fixed WIP limits are useful primarily as a signal to tell you when it's OK to activate new tasks from your backlog of new work. But the WIP limit can also obstruct the flow of work, especially when you have a mix of items with different levels of urgency.

So the key is to make sure you have a useful signal when to activate new work, and make sure the mechanism does not disrupt flow.

If you are setting up a workflow for an organization, you can accomplish this with a gatekeeper who performs triage on new tasks following agreed criteria, and sends the work that meets the urgency criteria along a special expedited routing. This allows the regular work to flow more smoothly. For example, engineers focus on the main development work, while the urgent problems are handled by a technical support team. This generates very high flow for both kinds of work and allows the teams to stay focused.

But for personal time management, it's the same person doing the triage and handling all the work, so we need a simpler process.

For me, it's been working great just to put the unfinished work on its own list, and to organize my day so that I spend most of my time and attention on that list. This creates a natural impetus to get all those things done. There is strong momentum and engagement.

Occasionally I go scan my list of new and incoming items. Because I am more attuned to all my existing WIP and focused on getting it finished, I have a stronger sense of how much capacity is available for taking on new stuff.

So I have essentially two signals for activating new work:
-- the non-urgent things just don't stand out till I have some available mental bandwidth -- if I am really busy with my current WIP, then these things tend to stay dormant
-- the urgent things that really can't wait stand out very clearly in contrast to those dormant items

This all happens very naturally without any special rules that could obstruct the flow.
February 11, 2022 at 5:44 | Registered CommenterSeraphim
Seraphim:

<< Maybe I am off track here, but often things like "lack of money", "lack of time", "lack of speed" are *symptoms* rather than causes. You might be able to find tips & tricks to make a marginal impact on the lack of X, but it doesn't fix the root cause.

The root cause is usually some kind of priority conflict... priority conflicts can also lead to overwhelm, which leads to task switching, indecision, etc. >>

This is a possibility. For what it's worth, I feel that my priorities are clear, as is my "future vision" (to borrow Mark's term from "Dreams"). But, as I said in my initial post, self-diagnosis is hard. Certainly, I don't have all the answers.

Sometimes my behavior appears to be classic procrastination. But more often, I start work fairly promptly but don't make much progress on it, as if I lack follow-though. It's the latter that leads me to describe this as a speed issue. That and the fact that the problem is most palpable when I spend 30 minutes to write what I think should be a 10-minute email, 3 hours to do an assignment that I think should take 1 hour, and so on. It's possible my expectations are unrealistic, but I don't they are wildly off-base. I'm not aiming for superlative performance; I just want to be moderately good. I concede that I do have mild perfectionism when it comes to written documents; that contributes to slowness.

While it would be easy to conclude that I lack "processing power," I tend to agree with you that I probably have the needed capacity (maybe most of us do). That processing power has shown through at times. It just seems to get "blocked" sometimes (resistance?) or doesn't always materialize fully.

Currently, I effectively have two part-time commitments. The workload is not onerous in either of them and supervision is limited. Lack of accountability might be an issue too, which might be resolved when I take on full-time work in the next stage of my career. I don't know yet.

Anyway, obviously, we don't need to put on a finger on exactly where I need to improve. And I don't want to overthink it. After all, there are plenty of concrete things that I can do that I haven't done before or haven't done in a while. (Mark's writings are a fountain of ideas.) I have optimism that I (re-)capture some of the efficiency in the near future. I only share the above by way of agreeing with you that these issues can be complex and possibly not exactly what they appear at first.
February 11, 2022 at 16:06 | Registered CommenterBelacqua
Seraphim:
<<So I have essentially two signals for activating new work:
-- the non-urgent things just don't stand out till I have some available mental bandwidth -- if I am really busy with my current WIP, then these things tend to stay dormant
-- the urgent things that really can't wait stand out very clearly in contrast to those dormant items
>>

I do not understand. Non-urgent things nor urgent things are not signals. Did you mean a different word than signals? Are you merely saying you look at your list and non-urgent things don't stand when you are busy out but urgent ones do?
February 11, 2022 at 17:16 | Registered CommenterAlan Baljeu
Belacqua: I'm very with you with the feeling that sometimes I don't operate nearly efficiently. If I'm caught in overwhelm and my choices are A B and C, I might do D instead. Or, I might do B but my mind is so distracted I complete it at much below my ordinary capacity. I think the approach to addressing such issues lies in finding clarity, something which I don't believe can be found in scanning a list of tasks.
February 11, 2022 at 17:45 | Registered CommenterAlan Baljeu
Belacqua:

I'll add another element to this discussion, specifically around "X should take this long, but takes much longer when I do it." Are we also sure that this is really an issue of efficiency? A lot of your assessment of speed seems to be centered around your conclusion that the discrepancy between your estimates and your actual time to completion results indicates a failure in efficiency. That *could* be the case in your specific situation, but speaking generally, that's almost 95% of the time *not* the case in most modern workloads.

There are a bunch of things in play [1]:

1. You fundamentally are making estimates of task completion times.

2. Accuracy of estimates is proportional to the clarity of not only the task outcome, but also the clarity of the exact physical operations required to complete that task as well as the sensitivity of the task to external, uncontrolled inputs (such as other people and data).

3. You will almost certainly estimate incorrectly, and you will almost underestimate, *even* if you take into account that you will underestimate (Hofstadter's Law).

4. Most [knowledge] work is not repeatable, so previous behaviors cannot predict accurate time estimates for future work.

5. Confidence in the feasibility of estimates leads to cognitive biases that prevent accurate diagnoses. (Estimates are not reality.)


The first thing to note is that you are fundamentally making estimates. A lot of people will make estimates and not believe they are estimates. In other words, they will assess their work, either before or after they have done it, and insist that a given task "shouldn't possibly take longer than X." Particularly here, the emphasis is the assertion that it *can* (that is, it is physically possible) be done in X time, leaving no room for doubt that these tasks at least theoretically are capable of being completed in a certain time frame. That's wrong, most of the time. This is because these time predictions are usually based on an idealized actor as well as an idealized definition of the work, both of which literally don't exist anywhere near reality. These predictions have to be taken as nothing but estimates, and that the estimates might not even represent a theoretical ideal. It may very well be the case that there exists no possible idealized state that could theoretically exist in a realistic world in which the estimate could have been met, and in fact, that's more likely to be the case than not.

Second, these estimates are often much more wildly inaccurate than people give them credit for. You don't have a good estimate of how your own time works or how long something will take, unless you are literally repeating the same exact sequence of physical actions each time, with almost no variability in mental processing required (such as in factory assembly work). Furthermore, in any complex work, you must engage in a series of ideas that must involve taking knowledge that you have not already put together and assembled and making something of that, whether collaboratively or internally, and that is a process that *cannot* be accurately predicted consistently without significant overhead. In other words, your tasks are *way* more variable than you give them credit for, almost all the time, including with regards to responding to email. Email is particularly interesting, because people will often enter a flow state when responding to a single email, or they engage in "sympathetic nervous system" responses. This means that they will lose their sense of time, inaccurately assess how long something is taking, and so forth, especially the more efficiently they are focusing on that email.

Third, it's almost a universal truism that everyone underestimates rather than overestimates, even when you try to compensate for this fact. You're almost always going to think that something should take less time than it really does.

Fourth, people have a nasty habit of thinking that two things are much more similar than they really are in knowledge work. Unless you are doing something that is exceptionally boring and repetitive, then it's very unlikely that you are doing the same tasks over and over again. So, for instance, each email is quite variable from the other, and the amount of time that it takes to respond or work up a decent email for some complex requirement is going to be highly variable, even when you restrict yourself artificially like saying that you will only allow yourself to respond with five sentences, or something like this. Put another way, accuracy of an estimate is inversely proportional to the value/complexity of the work. So, the more engaging and more interesting and more valuable the work is that you are doing, at least potentially, the less accurately you will be able to estimate it.

Fifth, people often look back at a thing or look at their estimates and incorrectly judge themselves to be inefficient or procrastinators or something else when something takes longer than they think it should. Holding tight to estimates and insisting that something should only take that long is not a good way to look at things. This is particularly true with individual work. A particularly nasty form of this is having a day in which you somehow get a bunch of things done at some speed, and then taking that as your baseline to expect to be able to repeat that performance. But that is, obviously, incorrect. That's because you won't ever be doing that same work again, most likely, and you won't ever necessarily be that same person again. You could have two projects or emails or the like which *seem* to be similar, but the chances of those two tasks actually taking the same amount of time is probably *way* lower if they are actually meaningful projects. If you find that you *are* able to accurately estimate how long they are taking, that means that you probably shouldn't be doing those tasks in the first place, and that they should be somehow delegated or outsourced or automated in some way that makes the inefficient repetition unnecessary.

Put another way, accurately predictable repetition in your daily practices on new work is a sign of inefficiency of the total system, even if it grants increased predictability. And chasing the ability to accurately predict performance is a red herring.

For example, when responding to email, there are many emails that people assess and respond to in the exact same way, such as in dealing with newsletters, spam, or scheduling tasks, or the like. These are highly predictable activities, and you might feel good saying, "Doing this sort of email is going to take exactly 5 minutes." But that's a sign that you're wasting time if you can actually meet that estimate. Instead, if the newsletter doesn't result in very interesting thinking or something of value when you read it (thus resulting in unknown amounts of time spent on it), you probably shouldn't even be getting that email in the first place. Scheduling can probably be automated, and all sorts of things can be more or less automated. Personally, I actually accept some intentional inefficiencies because I like them that way, but I don't mistake that for efficiency just because I can reasonably accurately predict that task, it's just a repetitive task that I like to do.

If you actually work to get low-value work off your plate somehow, then the amount of variability in your work will increase, and you will almost never be able to predict how long something is going to take. This is at least if you are in a creator role or knowledge work.


So, in conclusion, I want to be a little radical here, and presumptuous of your situation. You said, "It's possible my expectations are unrealistic, but I don't they are wildly off-base. I'm not aiming for superlative performance; I just want to be moderately good." I'll respond by suggesting that, indeed, your expectations actually *are* unrealistic and most likely wildly off-base. If you want to aim for improved performance, focus less on how much time you think something should take, and more time on doing the things that you think hold the highest value. Then, focus on engaging with those activities with as much clarity as you can and maximizing your flow state in each of those tasks. Gain that clarity through brutal incrementality and decomposition. Then, ignore the timer. If you're doing the highest value thing at the moment (whether that's taking out the garbage or writing an email or synthesizing a cure for cancer), and you incrementalize that sufficiently so that you can be sure that you are adjusting your "focus" at the appropriate intervals, then you shouldn't have to worry at all how long something takes. Let yourself be free and confident to lose yourself in the work.

[1] Benson, Jim. "Why Plans Fail: Cognitive Bias, Decision Making, and Your Business". 2011. Modus Operandi Press. https://www.amazon.com/Why-Plans-Fail-Cooperandi-Mememachine-ebook/dp/B006S3UHGA/
February 11, 2022 at 22:42 | Registered CommenterAaron Hsu
Alan Baljeu wrote:
<< Did you mean a different word than signals? >>

Sorry I wasn't very clear. Here is what I meant: the signal to pull work is the "standing out".

1-- Non-urgent things stand out only when I have enough mental capacity for it. When I am busy with lots of WIP already -- and I can feel the load very clearly because it's all in one place on my Unfinished list -- then the new work doesn't stand out. When it does start to stand out, it's because I have been freeing up some mental bandwidth and feel ready to take on more.

2-- Urgent things stand out even when I am already loaded up with WIP.

Maybe it's really only one signal -- the standing out. It just seems to work really effectively with this AF4R system because the Unfinished and Recurring lists are so good at giving a strong sense of my current load of committed work, compared to, say, AF1 or Simple Scanning, where all the different tasks are mixed together without any differentiation. With those systems, "standing out" can sometimes just mean "I feel like a break right now, and this task looks like a lot of fun". It's easier to maintain the illusion that I can absorb new things without any real negative impact. That illusion disappears when the Unfinished and Recurring are identified clearly on their own lists.

Does that make it any clearer?
February 12, 2022 at 3:05 | Registered CommenterSeraphim
Hi Aaron,

You make some powerful points.

I’m willing to believe that, in general, we’re probably not too accurate when estimating how long something takes.

I may have given the false impression that I’m obsessively estimating task completion times. Actually, I’m not a clock-watcher by habit. I was just trying to describe the (fairly common) feeling one gets when one finishes something and concludes: “That took too long.” Granted, if what you say is true, that thinking might not be especially helpful. Though I suppose we don't have the luxury of entirely ignoring time either.

Perhaps your post leaves as an open question whether an approach that would take the guesswork out of timing (e.g., time tracking, stopwatch) would have any value.

In your last paragraph, you suggest:

<< [F]ocus on engaging with [high-value] activities with as much clarity as you can and maximizing your flow state in each of those tasks. Gain that clarity through brutal incrementality and decomposition. Then, ignore the timer. If you're doing the highest value thing at the moment (whether that's taking out the garbage or writing an email or synthesizing a cure for cancer), and you incrementalize that sufficiently so that you can be sure that you are adjusting your "focus" at the appropriate intervals, then you shouldn't have to worry at all how long something takes. Let yourself be free and confident to lose yourself in the work. >>

Are you suggesting: Choose high-value activities, use timing/timeboxing extensively until you’re more efficient and your focus improves, then stop timing yourself? Or I have I misunderstood?
February 12, 2022 at 15:42 | Registered CommenterBelacqua
Fascinating discussion.

1) I don't agree that it's not possible to estimate the time something will take at least accurately enough to be useful. Like most things in life it requires practice. Something as simple as writing down an estimate and then timing yourself to see how accurate the estimate was will increase your accuracy quite considerably. Whether that's useful information or not will depend on the nature of your work.

2) Although it may not be possible (or even desirable) to estimate the time two pieces of work will take, it is much more likely to be possible to tell which is going to take longer. In other words estimating the relative time between tasks and projects is easier than estimating the actual time they will take.

3) The various methods of time-boxing are remarkably efficient at concentrating your work.
February 12, 2022 at 22:37 | Registered CommenterMark Forster
I haven't had a chance to read carefully the entire post, but I might just chime in that in my life the one thing that has made the most impact in the speed of my work is the length of gap between tasks (I think I saw that point in a list above), even between tasks within a task. Engineers might have called this "single minute exchange of die " (SMED) in manufacturing, but I think the principle applies. The idea was to be quicker at starting the next phase of work. An example: as a teenager, I used to pick cherries. I could fill a 6 quart basked in about 30 minutes - not much of a way to make a living doing piece work. But one of the older workers could pick one basket in 10 minutes - amazing. The difference, his work just seemed to flow. I asked him about it, and he said I was taking too long between "cherries." He used two hands, picked multiple handfuls quickly and didn't waste any time between the basket and the branch. What I found was that the effort in speeding up my work was a more intense concentration and present moment awareness. I could get it to 15 minutes per basket, which I thought was pretty good. I know this doesn't seem significant, but those seconds between "things" really really add up.
February 13, 2022 at 19:39 | Unregistered CommenterPaul MacNeil
Paul MacNeil:

That's a good example of how practice can shrink the time it takes to do a task (or increase how much work is done within a task). But it is also the case that practice will shrink the time between different tasks. The best tool for helping your practice in both cases is a stop-watch.
February 13, 2022 at 23:59 | Registered CommenterMark Forster
Hi Mark,

Is the basic idea with the stopwatch similar to track tracking: You see how long you take to do things and try to improve your efficiency? Perhaps with the benefit that having time tick away on the stopwatch display might provide an impetus to speed up?

I don't recall you discussing use of a stopwatch before, but I might be forgetting. I know you've written a lot about using a count-down timer (in Get Everything Done, etc.).
February 14, 2022 at 1:38 | Registered CommenterBelacqua
Belacqua:

<<Are you suggesting: Choose high-value activities, use timing/timeboxing extensively until you’re more efficient and your focus improves, then stop timing yourself? Or I have I misunderstood?>>

I think the problem is primarily with trying to predict how long things will take and with the "feeling" that things are taking too long. Neither of those is a reliable indicator of anything except your feelings and your impressions without more data.

I think timing and time boxing is valuable only once you give up on trying to fix/control both scope and time together. It doesn't matter whether it takes an hour or days to get something done provided that it remains the most valuable thing you can be doing at the moment. The principle is that it's much better to do the right thing more slowly than the wrong thing fast, and you can't really focus on going fast until you are going right.

Two highly successful methods in software development illustrate fixing one or the other but not both. Kanban says, you get to pick the scope (set of things we are working on right now), and you're limited to this number of them, but you don't get to tell us how long we have to get them done. They will take as long as they take to get them done. Extreme Programming says, we will work in fixed time boxes, and we will select some stuff to work on during that time, most important first. If something looks like it won't get done, we won't work harder or faster to get it done, we'll just not do it in that time box.

Thus, I think people are much better served by focusing on improving their ability to do one thing at a time (monomania) for extended periods of time before they try to do anything else. It's all about creating an environment for you to focus on a single thing (not a bunch of high value things), which will train focus and improve mental processing power. During this period of learning how to single task, I don't think there's any benefit to timing your work.

Once the above is done, you should then be able to construct a clear idea of the "flow" and process of you work. This is important. At this point, you can begin to actually work on optimizing that explicit process. You can track the cycle time of your work (Kanban), or you can use fixed time boxes (Extreme Programming). This gives you a way to gather statistics (including variance) on how long your work *actually* takes on average. This can allow you to make projections about probabilities of completion times.

Once you have that timing data, you can then try concrete changes to the process and evaluate their effectiveness. The important thing here is that you are never feeding into the idea of "this is taking too long." It's not about your feelings of how long something *should* take, but about how long it actually takes in practiced evidenced by historical data. You're still working at the same speed; you're just doing a different process. The important thing here, IMO, is a shift away from "accelerating my work speed" towards, "retaining a steady, sustainable pace for the long run," which is more valuable.

So, what I was really suggesting was to stop timing yourself at the beginning and only begin to consider timing after you have addressed the impediments to monomania.

Just as an example, monomania won't work if you can't adequately create the conditions for you to drive towards completion on a given task. That means that in order for you to succeed, there, you'll have to find some way to refine your tasks to drive to completion, which requires that you address any perfectionistic tendencies keeping your from defining a done state concretely and attainably. Many people, rather than addressing these issues, will just take them as static conditions of their lives that they have no control over and then just say that the "system doesn't work for me because...." That's a little like someone saying that they really want to get into their car and go for a drive somewhere, but they can't because they are barefoot and the path to the car is a painful, sharp gravel way. Well, there are any number of solutions to that. But if you suggest, "Well, go put your shoes on," they respond with, "But I'm barefoot, I can't put shoes on." [Aside: they could also just endure the pain, or they could hire someone to replace the gravel with something more pleasant or do it themselves.]
February 14, 2022 at 4:35 | Registered CommenterAaron Hsu
Belacqua wrote:
<< Is the basic idea with the stopwatch similar to track tracking: You see how long you take to do things and try to improve your efficiency? >>

I will be interested to see Mark's reply.

For me, using a stopwatch or a timer serves a few different purposes:

-- It makes me aware of how much time I am spending. I generally don't need to make a conscious effort to improve the efficiency -- the awareness alone is useful. This is similar to Mark's idea to observe how well you are feeling. He writes << Now, this is important: once you have given your answer do not try to make yourself feel better. Just carry on observing your feelings by regularly asking yourself “How good do I feel?” >> http://markforster.squarespace.com/blog/2017/7/25/the-most-important-thing-ive-ever-written.html

-- The timer helps me get started on large tasks that I am resisting. I just set the timer for 5-10 minutes and tell myself, just work on it for a few minutes, just try to poke at it and see what happens. That is usually enough to get me going

-- The timer helps me be more aware of time I am spending on time sinks. For me, this tends to be reading and researching. So I set a max time (typically ~30 minutes) and make myself take a break and do some other task after the timer stops.
February 14, 2022 at 5:46 | Registered CommenterSeraphim
Re stopwatches and timers: yesterday, I glanced at a couple of Google hits for "How to work faster." Most of the suggestions were forgettable, but one caught my eye:

1. Set a time for each task (you can predetermine this by timing past performance);
2. Use a countdown timer while you work;
3. Try to lower your times and set shorter deadlines.

It seems that it might be tricky to 1. Decide on an initial time goal; 2. Know what to do when you miss your target (Do you add a few minutes? Make it a new task/target? How do you avoid beating yourself up for "failing"?).

I don't know if this would be effective. It could run into the typical problem with unreal "deadlines" (you know there's no consequence for missing them). Aaron's concerns about estimating time also seem relevant here.
February 14, 2022 at 13:37 | Registered CommenterBelacqua
I use TickTick as my task manager. Whenever I start working on a task, I right-click the task and start the TickTick stopwatch timer. Immediately, a box pops up on the lower right of my monitor, listing the task name and the time it is taking. It also starts a "forest" sound which I now associate with focus. When I finish that task, I stop the timer. Then choose another task and so on. The really helpful part for me is that my TickTick daily calendar will list those tasks in time slotted order and ALSO show the gap between tasks. That, for me, is a key. I tend to spend too much time between tasks, either zoning out or having trouble choosing. I also track breaks using the same task method. I find it very interesting to evaluate my day this way.
February 14, 2022 at 15:16 | Unregistered Commentertomcal
Belacqua:

<<Is the basic idea with the stopwatch similar to track tracking: You see how long you take to do things and try to improve your efficiency? >>

This discussion is about speed, and if you were trying to increase your speed in track events you'd use a stopwatch, wouldn't you?

Here are some ways of using a stopwatch which spring immediately to mind:

1) time how long it takes from starting a task to finishing it.

2) time how long it takes from finishing one task to starting the next.

3) time how long it takes to write the first draft of an article.

If you keep records you can see how you are improving, which will tend to increase your speed in itself. Timing something also provides a much greater degree of concentration.
February 14, 2022 at 19:06 | Registered CommenterMark Forster
Belacqua:

To give you an idea of how much you can do if you teach yourself to work in a concentrated manner, using NQ-FVP I have just done 124 tasks in 24 hours (midday yesterday to midday today), which included two long walks. I am at "in-box zero" with all my routine tasks. And yes, I did get a good night's sleep - which wasn't included in the number of tasks!

Currently I have 78 tasks on my list, which at the present rate of progress is less than one day's work.
February 14, 2022 at 19:14 | Registered CommenterMark Forster
Mark Forster:

<< This discussion is about speed, and if you were trying to increase your speed in track events you'd use a stopwatch, wouldn't you? >>

Now that you point it out, it's seems strange that I haven't tried that before. I have tried countdown timers. And time tracking--writing the start time beside the next task (following your method in Get Everything Done). But a stopwatch is obviously different than a countdown timer. And it also seems subtly different from time tracking. It's another thing to try--thank you!
February 14, 2022 at 21:42 | Registered CommenterBelacqua
Mark:

<<To give you an idea of how much you can do if you teach yourself to work in a concentrated manner, using NQ-FVP I have just done 124 tasks in 24 hours>>

I've seen this metric before, but given your practice of using little and often and allowing tasks to be actioned but not completed count towards the score, I'm not sure I understand why this is a compelling leading indicator of valuable work done. I'd appreciate some more explanation about why you think counting the number of times you *switch* tasks, which is essentially what this is tracking, is a good metric.

Assuming your walks took about 1 hour total, and you got a comfortable 9 hours of sleep, that gives you 14 hours of work there (assuming you're not eating), which puts you at somewhere in the ballpark of 5 - 10 minutes on a single task. But say, essentially, "I spent on average 5 - 10 minutes on something before I switched to something else," doesn't really tell me much about how valuable you would consider that work at the end.

Just as an example, I'd say there are a set of daily things that I want to spend a specific amount of time on, then there are some things that have to be done to stay abreast of my life (administration), that I ideally want to spend as *little* time as possible on while still being 100% effective. Assuming those things are taken care of, then there are the set of tasks that are either enjoyable entertainment or valuable, impactful work. In these cases, it seems to me that it's more likely that there is an inverse relationship between my sense of accomplishment and the number of such items I would switch between or among within a day. That is, I'd rather spend 1 - 3 hours straight on something than a bunch of 10-minute chunks switching between a few things. In fact, the idea of not being able to spend hours or days at a time on a single thing or item that has my attention, but rather jumping among many different tasks and doing something "different" rapidly seems to me to be the absolute definition of a nightmare. And the more I am able to retain high focus and attention over a prolonged period of time at peak processing power, the better I'll feel.

IOW, I'm just not sure I understand how increasing the amount of task switching that you are doing sounds like a good idea. Even if I *liked* all the things I was switching between, it seems like torture.
February 15, 2022 at 4:29 | Registered CommenterAaron Hsu
Aaron:

<< I'm just not sure I understand how increasing the amount of task switching that you are doing sounds like a good idea.>>

I think you're rather missing the point of this discussion, which is about speed rather than value. One of the points made was the speed at which one could switch from task to task. Doing 124 tasks in 24 hours as you say indicates nothing about the value of the tasks. But it does illustrate that it is perfectly possible to switch from and to a large number of tasks extremely efficiently. This is a learned skill.
February 15, 2022 at 20:45 | Registered CommenterMark Forster
Allocate that as though it were a serious task - it deserves to be treated seriously, if you're going to be able to get a better balance in your life. It can't be all work and no play.
February 25, 2022 at 7:00 | Registered CommenterStephanie Sy
<< I'd rather spend 1 - 3 hours straight on something than a bunch of 10-minute chunks switching between a few things. In fact, the idea of not being able to spend hours or days at a time on a single thing or item that has my attention, but rather jumping among many different tasks and doing something "different" rapidly seems to me to be the absolute definition of a nightmare.>>

I have found that this is truly a personal thing, perhaps what makes us all neurologically diverse. I have a coworker that can sit and code for hours on a single project.. He accomplishes some amazing work. I, on the other hand, have a brain that likes to switch contexts often. As a result, I get to go to all of the meetings and get the dozens of inputs and requirements for the team and he gets to put the headphones on and focus for hours and rarely attend a meeting. He plays to his strengths and I play to mine. I'm learning that diverse teams are the strongest teams.

I also agree with Mark in that in the same way that "Deep Work" is something that we can learn and get better at, so is the ability to quickly and nimbly switch contexts. David Allen once compared this latter skill to a martial artists fighting multiple opponents at once. That's probably only a concept that happens in the movies but it illustrates the point.

Brent
February 25, 2022 at 23:53 | Unregistered CommenterBrent