« Is life is a series of ... wows? A selection of "made me think" ideas | Main | A conversation with Kerry Gleeson, author of "The Personal Efficiency Program" »

10 GTD "holes" (and how to plug them)

Since starting my study of the field of personal productivity (first GTD-related post: August 2005 - Actually Getting Things Done With Getting Things Done! Surprises And Learnings From My Implementation), I've been thinking about where the holes are in Allen's GTD and other modern systems [1]. I like the idea that there's a lot to learn in these interstitial spaces, no matter the object of study - for example art (in his book A Whole New Mind, Daniel Pink talks about seeing negative spaces [2]), social networks (seek opportunities to put yourself into structural holes - via Scott Allen [3]), and scientific discovery (Mendeleev and the periodic table, for example [4]).

Following are a few of the limitations I've noticed for GTD in particular. I wanted to share them with you and ask your thoughts on its shortcomings, and how to fix or work around them. In no special order:

Hard to sustain

One of the biggest challenges to people adopting new and better habits is getting them to stick. For a personal productivity practice (and I do mean practice, as in learning an instrument or studying yoga), we always face plateaus and fall off the wagon. This is nothing new - in my study of the field, this is a fundamental human limitation, so it's not fair to blame a particular author or method. That said, we need to structure work such that staying on the path (and getting back on it) is as easy as possible. In that sense, GTD is hard; there are many essential habits (the system is rather rigorous, after all), and lends itself to being brittle (see below).

For some ideas on how to deal with this, I'll point you to your own wisdom: Reader Question: Getting Personal Productivity Changes To Stick?

Not enough help with doing

Of course when it all comes down to it, we need to do the work we've so carefully collected and distilled. The question is whether GTD provides enough guidance on how to make progress. Just look around the productivity blogosphere and you'll find every tip imaginable around how to get ourselves to work (esp. procrastination [5]), including:

  • Time blocking/mapping [6]
  • Covey's importance/urgency matrix
  • Worst first
  • Easiest first
  • Reward yourself
  • "Dashes" (e.g., (10+2)*5)
  • 80-20
  • Current initiative (see Do It Tomorrow)
  • Swiss cheese
  • ...

You can argue that having a master list of tasks isn't meant to answer the "How," just the "What." But still - it's clearly an area any self-management system should address.

Any favorites you've found helpful?

One big list too overwhelming

Depending on the job, GTD's master task list can have hundreds of items on it. Even grouping them into people/places/things needed to do them (i.e., "contexts") leaves dozens of to-dos per list. One could argue that every one of those items is the result of a "yes" we've said, but it's still overwhelming, and can lead to paralysis.

So how do we fix this? One practice I've found helpful is to create a daily plan. You can get details in Are Daily To-do Lists And GTD Compatible?, but the gist is to create a temporary list of work you'd like to accomplish today (tomorrow, if you're planning the night before), then use that to focus during the onslaught. The trap is switching over to the temporary list, and letting the master flounder.

Another approach is to use a more limited time horizon for what we put on the lists, i.e., how far out we're looking when we promise ourselves we'll take action. It's up to you, but think in terms of months, say 60 days. Of course you still need to track the stuff you know is coming up, say by keeping a "not doing" list.

Finally, distributing tasks into your calendar will manage overwhelm. The only work you see is what's on your calendar for today. There's still a big list, but it's watered down over time. There are some serious pitfalls to doing it this way, though, including "copying forward" work you didn't get to, which focuses on what you didn't get done.

Anyone doing any of these?

Too brittle

A client came up with "brittle" to describe how catastrophic falling out of practice can be with systems like GTD. Sometimes one whirlwind week is all it takes to unravel the thing, which is a huge limitation. So: What contributes to brittleness? I see two factors: A workload that hovers around your "maximum" setting, and an excessive stream of incoming items. The first factor means you have no buffer for work to grow and shrink. You are oversaturated. (Sadly, unlike a supersaturated solution, adding heat will not allow putting more in.) Sadly, trying to do more is a common "solution" we come up with to managing ourselves, along with sleeping less, exercising less, and spending more time at the lab.

The second factor (too much incoming) also shortens your buffer. If the volume is equal to or exceeds your ability to empty it, you'll fall behind. A good test: How long does it take to empty your inboxes on Monday morning? If you say 1/2 a day, is that too much?

I'd love to know your thoughts on this one.

Too complex

A common complaint about Allen's work is that it's just too complicated. There are two aspects of this: His overall phases that frame our work (see my What A Difference A Framework Can Make), and his diagram for deciding and emptying our inboxes. Now, since many of my clients [7] are from scientific and engineering fields, adopting such a strong, process-oriented scheme is a natural fit. But the critique is valid - I think it's one of those "it's so simple now that I've practiced it for two years" things.

To simplify the framework, any ideas? Here are a few variations:

  • Capture, Plan, Act - From an OmniFocus demo movie.
  • Think, Do, Enjoy - From SuMMy's comment on Extreme GTD: How Low Can You Go (or: Can We 80-20 GTD?):
    I agree everyone needs their own system my perspective is different though- there is a base of knowledge that people should start with and build on it using the pieces needed for themselves. The key is start with basic building blocks: Think (goals/planning/long term thinking), Do (processing rules), Enjoy (missing from gtd, it gives motivation and meaning to what needs to be done).

To simplify the deciding and emptying steps, there's a bunch of approaches, all with the same idea - make decisions about each item in order, then move it into the appropriate tracking bins. For example:

In addition to my own exercise in simplifying GTD (see Extreme GTD: How Low Can You Go (or: Can We 80-20 GTD?), you might enjoy Gina's thoughts on whittling down essentials: Practicing Simplified GTD and Email: Empty Your Inbox with the Trusted Trio.

No time use analysis

Almost every book I've read on time management and personal productivity starts with an analysis of how we use our time, often called a Time Log (see, for example, Jasper's book and Laura Stack's Leave the Office Earlier). However, does David Allen cheat by skipping this step? His perspective seems to be one of presenting a solid approach in detail, and assuming it's generally applicable to all kinds of work. It's also related to his bottom-up approach (see below). Peter Drucker in particular emphasizes this (e.g., "If you can't measure it, you can't manage it" - see What's Your Feed Reading Speed?).

The basic idea is to track for a representative number of days how you spend your time, analyze where the opportunities for improvement are, then adjust accordingly. That's why Mission Control's workshops (for example) start with this (see A GTD-er's Perspective On Mission Control's "Productivity And Accomplishment").

Variations on this problem: No interruption analysis. To do it, simply track for each interruption: Who interrupted, What was it about, Value of the interruption, and the Duration. Optional: Time of day.


No built-in balance

This is a big one: GTD has no built-in self-correction of work added vs. work performed. It enables analyzing and correcting imbalances, but doesn't address directly the fact that we can't add work without managing the consequences. It's like eating: You want cake every day? Then expect your weight and health to be impacted. Mark Forster's book Do It Tomorrow and Other Secrets of Time Management has a strong component around this: closed lists, lists that can't grow without removing something. Interestingly, a calendar is an example - hence the "schedule everything" alternate to to-do lists. A secondary problem is feeling overwhelmed by large lists (see below).

Related: When Inputs Exceed Your Workflow System's Capacity, What GTD And Weight Watchers Have In Common

Not goal-driven

Here's another common complaint: Allen starts "bottom-up," i.e., getting on top of our day-to-day workload before we consider higher-level aspects of our lives. This is in opposition to approaches that are goal-driven: Start with the important aspects of your life, then drill down through vision, goals, and projects to (finally) actions. The reasoning here is (to use Stephen Covey's analogy [8]) it makes no sense to climb a ladder if it's leaning against the wrong wall.

For me, starting top-down didn't work. I needed the "free your mind" piece before I started thinking about purpose. (And boy did it work - see Commitment Time! (Taking The Big Leap).) Interestingly, one thing that comes up in my interview series with top productivity experts is that the top-down/bottom-up issue is not all or nothing. Both components are important, and there's often a cycle or spiral that emphasizes work on one then the other over time.

The potential problem is this can lead to a disconnect between tasks (actions) and goals (purpose/motivation) - it's an important dimension besides actionable. For one possible solution, see Where Are You Going? Use Your Actions And Projects To Reverse Engineer Your Goals.

How do you connect goals with action?

No built in planning/task estimation

Central to many work schemes is the notion of using estimation and measurement to inform action choice and project planning. In You Need to Get to Work!, Julie Morgenstern says the most important skill is answering the question "How long will it take?" and then learning to "accurately and honestly estimate it in advance." Without this, it's very difficult to plan methodically - which is unneccesarily stressful. As Mark Forster says in One Thing at a Time:

Don't completely erase the old estimated completion dates when you revise them. That way you keep a record of how many changes you have needed to make. Examining that record can tell you a lot about your workload and the way you are tackling it.
Interestingly, in the Extreme Programming methodology [9], tracking how long work takes is essential to adjusting course. After all, how can one predict how long adding a feature will take if you have no measure of past performance to go on?

Two resources you might like: In To Do Doing Done, Snead and Wycoff share different ways to estimate, including asking someone else how long it took, or using more formal equations. And Jose has a nice little discussion at Bias and Accuracy in Estimates of Task Duration using Academic Tasks.

What do you think - is building in tracking and estimating something a productivity method should do?

No specific accommodation of personality types

Along with not doing an initial time analysis, Allen also doesn't do an assessment of personal work style. This is another common starting point in other time management books, and I wonder whether Allen is ignoring the problem, or whether there is no problem, i.e., GTD applies to everyone, and personality doesn't matter. The thought is that we all have different personality types (e.g., Myers-Briggs) and this should inform how we approach work. Someone who is more "divergent" may need a workspace very different from "convergent" types (vocabulary from Time Management for Unmanageable People). This information might also be useful in deciding a starting place. Here's a list from Morgenstern's Never Check E-Mail In the Morning: What's standing in your way? Is it you or is it them?

  • You don't plan well.
  • You lack confidence on some tasks.
  • You are unable to prioritize.
  • You're a perfectionist.
  • You feel guilty saying no.
  • You gravitate toward quick, easy tasks.
  • You're poor at estimating how long things take.
  • You're physically disorganized.
  • You start many things, finish none.

What's your take on this? Is it important to account for personality? Or is it that these days it's not possible to sell a book that requires too much up-front quiz-taking and assessment, due to shorter attention spans?


  • [1] Here are a few other systems, from my AP post answers to your academic productivity questions:
    Could you recommend other systems or methods besides GTD, or is GTD the best thing since sliced apple pie for academics? ... Mark Forster's Do It Tomorrow (my interview with Mark is here), Chris Crouch's Getting Organized (interview here), Sally McGhee's Take Back Your Life! (interview here), and Kerry Gleeson's The Personal Efficiency Program (interview here).

  • [2] Pink describes negative spaces as "the part of the big picture we often overlook. So train your eyes to see it. When you take a walk or browse a store or page through a magazine, peer past what's prominent and examine what's between, beyond, and around it. Being aware of negative space will change how you look at your surroundings - and it will make the positive space snap into focus. It's also a way to be surprised."
  • [3] Allen defines a structural hole as "the weak connection between two clusters of densely connected people" From Structural Holes: The Social Structure of Competition via The Virtual Handshake: Opening Doors And Closing Deals Online.
  • [4] From Dmitri Mendeleev's Wikipedia entry:
    Initially, Mendeleev was derided for there being gaps in the table. Ultimately though, he was vindicated when previously unknown elements (notably scandium, gallium and germanium) were discovered that filled in these holes and possessed properties (atomic weight, density, melting point, etc.) close to what Mendeleev predicted.[3]

  • [5]Brian Tracy lists a number of them in Time Power, including:

    • think on paper
    • gather all the materials and work tools that you will need before you begin
    • do one small thing to get started
    • "salami slice" the task
    • practice the Swiss cheese technique
    • start from the outside and complete the smaller tasks first
    • start from the inside and complete the larger tasks first
    • do the task that causes you the most fear or anxiety
    • start your day with the most unpleasant task first
    • think about the negative consequences of not doing the job or completing the task
    • think about how you will benefit from doing the job or completing the task
    • set aside fifteen minutes during the day when you will work on your project
    • resist the tendency toward perfectionism
    • pick one area where procrastination is hurting you
    • develop a compulsion for closure
    • maintain a fast tempo

  • [6] Julie Morgenstern popularized this term. See this newsletter article. From Organizing from the Inside Out:
    [a time map] allots specific spaces in your schedule for tending to the various core activities of your life. It serves as a foundation from which to work that forces you to keep your life in balance, giving you all the time you need to accomplish your goals.

  • [7] As usual, I want to be very clear that I have no association with David Allen or his company. His work has been a big influence, but I continue to combine the best practices from many sources (which I share here) into my work.
  • [8] You might enjoy Leo's: Exclusive Interview: Stephen Covey on His Morning Routine, Blogs, Technology, GTD and The Secret.
  • [9]You can read a bit more on XP in Is GTD the "Extreme Programming" of Time Management? and Productivity for Programmers, #2: Efficient vs. Effective.

Reader Comments (35)

One word: Wife.

June 3, 2008 | Unregistered CommenterAnonymous

The Kurt Lewin Change model, although it sounds very simple, is very effective for managing change. It simply goes thus:

Unfreeze - Change - Refreeze

Simply put, it means getting people out of old unproductive habits, make the change, and reinforce new activities, till they become a habit or the equilibrium state.

We implemented a new web based productivity and intranet software called [ HyperOffice | http://www.hyperoffice.com ], which included many tools like to do lists, intranet workspaces, document storage and remote access, calendars etc, which we were sure would help improve productivity. The challenge was to habituate users to using it so that it would become an integral part of their day to day day working. The Kurt Lewin model really helped us think of ways to "refreeze" this change.

October 31, 2008 | Unregistered CommenterPankaj

Much appreciated. Cool app, too.

November 2, 2008 | Unregistered Commentermatthewcornell

Good points all of them. Just two things I want to comment on: first while GTD is quite brittle, I also find it's easy to pick up the pieces after it's broken.

Second, I think you forgot the most frustrating aspect of GTD: there is no good way to handle due dates for actions. This is one thing that I have yet to find a good solution to.

April 14, 2009 | Unregistered CommenterWellington Grey

Easy to pick up the pieces: True. Re-induction is a mechanical process that mostly involves time, once the concepts are in place. And a brain dump should capture a good bit of the "hot" activities. Nice point.

Due dates for actions: I've been thinking about this too. It's a common variation/extension, but I haven't totally wrapped my head around it yet. For a project time-line I use milestone reminder entries/dates. Dates for individual actions can be used as filtering criteria in more complex apps, which allows a more sophisticated deferment of action than Someday/Maybe. Then again, a reminder would do just as well. I suppose it's consistent to have all actions in the same list...

Thanks, Wellington.

April 14, 2009 | Unregistered Commentermatthewcornell

In my system I can optionally assign due dates to next actions. These automatically float to the top of the Next Actions view and show up in the exported calendar feed.

What's tricky is dependent actions. If I have action B that depends on A, and B has a due date, do I set a due date for A?

April 14, 2009 | Unregistered CommenterBrock Tice

So it's primarily for filtering actions?

April 14, 2009 | Unregistered Commentermatthewcornell

Hey, this is a great post. GTD is an amazing way to stay on track and be productive but there are still going to be holes. Thanks for addressing this issue!
We recently did a webinar with David Allen that you might find helpful! He touched on some key topics around GTD and answered some great questions. Check out the recording here: http://bit.ly/bOyk1o

August 17, 2010 | Unregistered CommenterAlanah Throop

Your GTD tips are really good. Another way to properly manage your GTD is also using [ gyration drivers | http://www.driversupdate.org/gyration.htm ].

Thank you for your webinar, Alanah. The interview is indeed very helpful and brings loads of answers.

September 9, 2010 | Unregistered CommenterAlice W.
Hey Matt,

This is a great analysis of the GTD approach. The self-improvement genre in general has many holes because a lot of it comes down to the individual rather than the method - can the individual stick with the recommendations and does he have the commitment and will to break the bad habits. (this is kind of your point in 'Hard to Sustain')

But, I think most of your critiques are very accurate. I think for me the 'No build in planning / task estimation" was the biggest gap. To make up for it, I have started to use my calendar more often and planing things instead of just jotting down a task.

There are also many free applications and even browser extensions - I use "To Do List" for chrome.

I also made this infographic that helps conceptualize the GTD method: http://www.liquidplanner.com/blog/infographic-how-to-manage-productivity-to-get-things-done-gtd/

August 9, 2013 | Unregistered Commenternick nielson

PostPost a New Comment

Enter your information below to add a new comment.

My response is on my own website »
Author Email (optional):
Author URL (optional):
All HTML will be escaped. Hyperlinks will be created for URLs automatically.