knowledge work

Management Thinking, Productivity, Uncategorized

Getting Knowledge Work to Done

Mike Cohn at Mountain Goat Software takes on the idea of Getting to Done.

Cohn’s insightful post is focused on scrum as a software development process but applies to self-management and knowledge work. Let me explain how.

Cohn’s post considers four ways that completing work helps us. He defends the position that it is better to finish five tasks and leave five unstarted than it is to make partial progress on ten tasks. Here are the four ideas. I’ve changed the emphasis of the fourth and added a fifth:

Finishing gives faster feedback. In delivering software applications, feedback is critical. We get an idea what users want and try to build it. We think we’ve done a respectable job, but never really know until the users get their hands on that part of the application, use it, and let us know. Knowledge work is similar. We never really know how good a job we’ve done until we implement the new or improved process and measure the results or deliver the completed report and get feedback. Neither of these can be done with a partially finished knowledge work task.

Finishing gives faster payback. In software, an unfinished feature is similar to inventory and work-in-progress in a manufacturing plant. We have paid for the inputs, but cannot yet sell the product. Until we finish, we won’t get paid and we can’t cover the cost of the inputs. In knowledge work, we have used up some of our valuable attention, and perhaps our employer has paid for our time, but we don’t have anything that we can implement or deliver to recover those costs until we have finished the task.

Estimating progress on unfinished work is a hard problem. Software is well known for the “90% complete” problem. Software project managers have a running joke that programming tasks are “90% complete for 90% of the time”. Cohn’s post makes the point that this could be literally true in the programmer’s mind because as we make progress on a task, we find more detail about what needs to be done to actually complete the task. I believe this to be true of complex knowledge work tasks as well. Regardless, we are terrible at estimating how long things will take, particularly when the task is something we have never done before. On the other hand, we are great at knowing a task is unstarted, and decent at knowing when it is finished. In between, we are terrible at estimating how far we have come and how much we have left to do. And (see the next point) the more partially finished tasks we have, the worse we are at figuring out when they will all be done.

Work should be “not started” or “done” with nothing in between. Here, I’ll emphasize a different point than Cohn’s post. While we are working on a product, a software feature, or a knowledge work task, it is in a very complex state. A finished task is conceptually very simple; an unstarted task is slightly more complicated, but still relatively simple. A partially finished task is the most complex by far as we must keep track of the moving parts within the task.
Imagine a factory with lots of work-in-progress. When we stop work while Product A is unfinished, we must put it away somewhere, clean and reset the tools, and get out the plans/instructions for Product B. When we restart work on Product A, we must find where we put it away, figure out where we left off, take it to its next machine, and reset the tools. Only then can we get it finished and shipped. All these tasks are unproductive, and thus waste.
There is similar waste in “reloading” a complex knowledge work task. I’ll illustrate with a complex writing task. Imagine we’ve set aside a partially complete version of Report A to work on Report B. When we return to work on Report A, we must complete several steps just to get ready to move it along. We must find the right version and figure out the page where we left off. We must get out our supporting materials or refind a web page we are referring to. We must get into a similar mental state as we had when we left off. I call these steps “reloading” the task. This reloading may take seconds for a very simple task, but can take several minutes for a complex task. Also, reloading is a cognitively demanding, mistake-prone task, so it uses up some of the brainpower that we could use to complete the report. Finally, we know reloading is hard, so it gives us another reason to procrastinate and turn to simpler, less important work.

Finally, I’ll add a new point: The psychology of starting. We feel somewhat relieved when we’ve started a task. Depending on our work culture, reporting that we haven’t started may be associated with ignoring work. Therefore, we feel bad about not starting. So, it may be challenging to adopt the “not started or done” mindset. However, I guarantee that, as a boss, I’d rather have something complete than nothing. If you deliver five complete out of ten, at least I can move five projects along. If you deliver zero complete out of ten, I’m helpless. In addition, over time, the “not started or done” mindset will help your boss think hard about the priority of individual tasks. This will get the whole team working on the problem of having multiple number one priorities.

The expenses and challenges associated with work-in-progress are well known in the manufacturing world. Cohn’s post shows similar challenges in the world of agile software development. I hope I have made the point that work-in-progress is also expensive and challenging in everyday knowledge work. Get finished, or don’t start. Devote your attention to a single task, Get to Done what you can, and deliver it, even if some of your tasks make no progress.

1 Comment
Related posts
Time According to Drucker’s Watch
November 19, 2016
6 Reasons Planning Is a Pain (and Why You Should Do It Anyway)
November 9, 2016
Are You an Overwhelmed Leader?
October 10, 2016
Books, Management Thinking, Productivity

Time According to Drucker’s Watch

qtq80-KJjZNvI just finished reading The Essential Drucker. If you don’t know who Peter Drucker was, you probably should, because he was a leading thinker on organizations and management. He also coined the term knowledge worker more than fifty years ago. He wrote about how the advent of the knowledge worker would change the nature of executive and managerial work, along with the structure of organizations. Drucker – Wikipedia

I find all of his writing instructive. The Effective Executive (for example) is a 50-year-old treatise on the job of an executive in the modern corporation and one of the handful of nonfiction books that I reread occasionally.

Essential includes a chapter titled “Know Your Time,” which describes practices for time management. While decades old, it is particularly useful. In this post, I’ll summarize the main points.


Time is a unique resource. It is completely perishable. In the “accomplishment” process, there is no substitute for time as an input. It takes an instant to have the “aha” solution to a problem, but it takes time to get to the point of understanding the problem well enough to have the “aha”.

We are terrible judges of how we have used time in the past. Invariably, we need a time log, an audit, to tell us how we have spent our time over the last few weeks or months.

The time demands

Much of what we spend time on is unproductive, even if it is work. Our organizations require communication, which is rarely done efficiently. We must build relationships with each other, which cannot be done in an urgent manner.

However, knowledge work requires large chunks of time. Creating a presentation from scratch needs a fairly large chunk to get started, then editing can be done in smaller time increments. Strategic thinking and problem solving need large chunks. Getting work done through other people, the primary task of an executive, requires chunks. Any deep work, by definition, needs sizable chunks.

Time diagnosis

While we have been recording time for manual work since the start of the 20th century, we do not normally record time for knowledge work. But time is more important for the knowledge worker than it is for the manual worker. It is simply easier to record for manual workers.

So, effective knowledge workers track their time. The details of the method are not important, but the act of recording time is the first essential step in analyzing how we use it. Time tracking needs to be done for fairly long periods, weeks or a month, in order to account for daily and weekly variation. And our time drifts toward trivia, so tracking needs to be done routinely, at least a couple of times per year. Once we have recorded our time use, we analyze it for three areas of time use:

  1. Identify and eliminate things that don’t need to be done at all. There are probably more of these things than your memory will let you see. Some thought is usually required as our default is to be lazy and simply declare that things “have to be done”. It is fairly easy to test the assumption – stop doing it for a while and see if anyone notices.
  2. Identify and assign things that could be done by others. This is not exactly delegation, but a complete giving over of authority to another person. Again, we have to be tough on ourselves here as the lazy way out is to default to “I have to do it”. Perhaps the task needs to be tweaked or better automated before we completely hand it off.
  3. Identify and eliminate things that we do that waste others’ time. The only way to do this is to ask the honest question of our peers and direct reports. As leaders, when we ask for reports and status meetings, we must be ruthless with ourselves about the need for them. Again, our default is the enemy here. People likely need less oversight than we think and will ask for it if we don’t offer.

Pruning the time wasters

Having dealt with the things we have individual authority over, we now turn to those things that are organizational in nature. Organizational misalignment and poor structure waste everyone’s time, but primarily the time of leadership. Now we analyze our time record for four areas of concern:

  1. Identify the “routine crises”. These represent a lack of system or a lack of foresight. After the second occurrence of a crisis, we should assume that it will become routine if we don’t do something systemic to prevent it. Budget cycles, seasonal sales, release cycles, and recurring external events, such as employee turnover may trigger these broken processes that need to be fixed. Well-designed work and organizations should normally lack drama.
  2. Over-staffing wastes time. In an overstaffed organization people collide with each other too often. This results in jurisdictional disputes, politics, and other signs that people are in each others’ way. Knowledge workers need to be able to do their work with a minimum of explaining to others and asking permission. If the leadership of the organization spends more than a small fraction, say 10%, of their time on these kinds of issues, overstaffing may be the cause.
  3. Poor organization shows up in excess meetings. Meetings are caused by a need to exchange information. In a perfect world, everyone would know what they need to know to get their job done. However, we live in an imperfect world. Nevertheless, good meeting discipline requires severely limiting the number of meetings. I would add that some meetings may be caused by having hired staff that are unequal to their task and consistently require the help or advice of others to get anything done.
  4. Poor distribution of information wastes time. Routine calls for information and “bystanding” in meetings to see if there’s “something I need to know” can be eliminated by setting up proper distribution channels and creating meeting discipline.

Solutions to these seven kinds of problems may be obvious once they are noticed. Or they may require a long, disciplined effort. But the result of this work is useful for the entire organization, especially in terms of time recovered.

Consolidating discretionary time

The key to consolidation is to move things around to create the needed chunks of time. The problem is that this is a perpetual need. Non-discretionary time expands; new meetings are called; new crises pop up. Therefore, effective knowledge workers need to be consistently vigilant over their schedules.

Finally, it is difficult to know when knowledge work tasks, done during the chunks of discretionary time, are finished. We all have perfectionist tendencies and overdo. If we create 10 hours of discretionary time per week, but use it poorly, we are really no better off. Think in terms of deadlines for discretionary projects. Time box efforts and stick to the time allotted. Understand what is the minimum effective product and move on when it has been achieved.

I would add that attention management is a requisite skill in time management. If we can’t stay focused on tasks we set ourselves during discretionary time, we are not using it well. After, perhaps, decades of fragmented attention use, our ability to focus intently for periods of time may be weak. Don’t forget to defend your attention when entering into periods of discretionary time.

Time is more scarce than any other commodity. So it should be used more carefully than any other. The first step is a time audit.

Question: Have you ever audited your time? If so, let us know one key insight you gleaned. If not, perhaps now would be a good time to start. If you do, come back and share your insights.

Leave a comment
Related posts
Getting Knowledge Work to Done
May 6, 2017
6 Reasons Planning Is a Pain (and Why You Should Do It Anyway)
November 9, 2016
Were You Trained as a Knowledge Worker?
October 1, 2016
GTD, Management Thinking, Self Improvement

6 Reasons Planning Is a Pain (and Why You Should Do It Anyway)

qtq80-WsNgkaSmart, capable people often struggle with one crucial aspect of knowledge work: planning. In my experience, clients resist planning for a number of reasons. Here are some of the most common.

Planning causes mental pain, at least at first. Planning is taxing work. If you haven’t done it recently or it’s not a habit, it will literally cause pain when you do it. My university students helped me define this pain; it is the feeling of frustration. “It’s too hard.” “I’m just not good at this.” So, we give up too easily, avoiding the frustration of learning how to plan and developing the habit of planning.

Planning takes time. On average, your weekly planning will take 45 minutes or so. That’s for a routine week. If your upcoming week is challenging, planning may take twice that long. Daily planning will take 10-15 minutes. As with many activities, it feels easier to just jump in and start work. Doing so makes us feel that we are, at least, making progress. So we skip planning and just get started.

Planning is deep work. Some people think that planning is merely running over their to-do list. But real planning is the difficult exercise of estimating how long tasks will take and a thoughtful look at our tasks versus our constraints. It forces us to clearly define our commitments and deliverables. Most of us prefer to avoid this deep work, so we substitute a shallow, watered down version. We leave our commitments murky and don’t clarify through planning.

Planning forces us to say “No”. We don’t like to say no to others. Good planning shows conflicting commitments and when our capacity is full. Thus, planning forces us to make choices that likely result in difficult conversations. This brings us face to face with our priorities, which may not be clear. It is much easier, in the short run, to play hero and worry about the consequences later, so we don’t plan.

Planning shows hard truths. When you plan, you realize your limitations and understand your constraints. This is much more challenging than just believing everything will work out. Over time, planning also shows you how long it actually takes to get your tasks accomplished. This can challenge your mental image of how good you are at some of your tasks. These challenges are uncomfortable; we avoid planning so we don’t have to face them.

Planning is inexact. Even the most experienced planners and project managers cannot predict the future. Things happen. Key resources take a sick day. A supplier’s truck breaks down. In addition, we are predictably bad estimators, typically underestimating both the time it takes to do something and how much discretionary time we will have in a given day or week. Our plans don’t work out, so we abandon planning.

So, we resist planning because it is painful, time-consuming deep work, that doesn’t work perfectly and forces us to confront an uncomfortable reality. But I would like to show two essential parameters that planning produces: The first is time boxes that show us how much time we can allocate to any given task. The second is milestones, which show where we should be on a multi-step project. Clarity on these two parameters is our responsibility in managing our schedule and tasks.

Time boxes force us to deliver even when the level of quality is unclear. We can be perfectionists and let deadlines slip chasing perfection. Remember that the quality of knowledge work is notoriously difficult to define. Challenge yourself to work hard for the duration, but, when the time is up, ship the result. It is probably at least good enough. Time boxes help give us the courage to “just ship it” on the deadline.

But, even if it is not good enough, you probably don’t know how to fix it. My experience as a writer shows this challenge. I invariably think part X is bad while part Y is good. However, my reviewers and editors think the opposite. So, rather than continuing to perfect part X, I should have been working on part Y. The same holds true for much of our work: presentations, reports, and communication. The parts we think need additional work are actually OK, but we overlook parts that need shoring up. So, instead of exceeding our time boxes and blowing our schedule by chasing our view of perfection, we should set shorter time boxes and use the additional time to get feedback on what needs attention.

In addition, using time boxes gives good historical data on how long a task actually takes. Working on a deliverable in short spurts is inefficient and makes it difficult to tell how much work we actually did. We need to know how long the task took last time in order to guess better how long it will take this time. Disciplined work during time boxes shows us that.

Milestones are advance warning signs that a project may not turn out well. Milestones give us time to renegotiate our commitments while our stakeholders still have time to adjust. In my experience as a project manager, people are usually reasonable when they know of a problem in advance. They are less reasonable when they find out today that we don’t have something they expected to get yesterday. Some of our work is speculative; we guess, in advance, how long a task will take. Sometimes those guesses are wrong. Maybe our stakeholders can do without a piece of the work. Maybe we misunderstood their deadline. Maybe they’re happier with a blue one on time than a green one that is late. But, we only find out by having the conversation. Milestones help us do that earlier.

Planning is challenging and can be painful. But, good planning produces time boxes and milestones. These parameters help us face reality, rather than living in an imaginary, but happy, world of false bravado and superheros who can overcome any constraint. They parameters help us keep our time under control and give us the best chance to make our stakeholders happy. Planning is imperfect, but, when it comes to executing our work, an imperfect view of reality beats a 20/20 view of a fantasy every time. So, overcome the pain and challenges and plan your work.

Question: What is your biggest pain point when keeping on track with your daily and weekly planning sessions? How much does it interfere with creating the habit?

Leave a comment
Related posts
Getting Knowledge Work to Done
May 6, 2017
Time According to Drucker’s Watch
November 19, 2016
Were You Trained as a Knowledge Worker?
October 1, 2016
Management Thinking

Were You Trained as a Knowledge Worker?

Like it or not, your work world, and the work world of the next generation, is changing.  And I’ll bet you were not trained for it.

Current work and future work are founded in knowledge work.  As Peter Drucker taught us in the 1960s, knowledge work is different from the routine work of 60 or 75 years ago.

  • It engages our brains more than our bodies.
  • It involves working in teams in interdependent ways.
  • It is difficult, if not impossible, to manage knowledge work in traditional ways.

I want to explore this third aspect of knowledge work further because it has subtle implications on how we need to approach work and personal work management.  I’ll start by explaining three of the assumptions of traditional management.  I’ll focus on one of them and it’s implications on our self-management training and practices.

Traditional management was designed to manage routine work.  Routine work is observable, thus traditional management relies on observation.  Factories were designed to make routine work tasks as independent as possible, so teamwork was not required.  Managers who designed the factory knew best how to perform routine work within it.

All three of these assumptions have important implications on how we need to approach work in the 21st century, but I’ll focus on the third in this post – the boss knows best.

Traditional management is built on the assumption that the higher you rise in the organization, the better you understand how to solve organizational or customer problems.  As you rise through the traditional organization, you learn more broadly about the firm and it’s operations. On the other hand, the lower your position in the hierarchy, the more you get told what to do and how to do it.  From the perspective of the worker, this aspect of traditional management can be both annoying and comforting.

The annoying part is constantly being told what to do and lacking freedom. Being under the control of another person is a challenge for anyone.   Every worker has, at one time or another, accused the boss of “micromanaging.”  This is a typical response when the worker disagrees with how the manager wants the job done.  However, in traditional management, the boss actually had a claim to better knowledge and the employee was more successful when doing work the way the boss wanted it done.

The more subtle, comforting part of this tacit arrangement is that the worker does not have to take much responsibility for the results of actions they were told to do, decisions made by the manager, and schedules set by the manager.  After all, the best excuse for a problem is “you told me to” – just ask my kids.  In the former economy, many workers simply didn’t have to take much responsibility for outcomes.  This is, in some ways, a comforting thing for the worker.

Now that we have moved into a knowledge work economy, and organizations are pushing authority down the hierarchy, the annoyance of being under the control of the boss is decreasing.  Workers have increasing freedom on what to do, how to get it done, and when to do it.  To the degree that workers are aware of this change, they usually celebrate their increasing freedom.

But the comforting part, having little responsibility, is also changing.  Increasingly, workers are being given responsibility for the outcomes of their actions and choices.  To the degree that workers are aware if this change, they are less thrilled about their increasing responsibility.

This change in freedom and responsibility in the workplace represents a challenge to our self-management processes and habits.  Firms are increasingly expecting employees to be able to manage themselves.  However, most of our current workforce has not been trained in self-management.  Think back on how well you were trained in this area…

  • Were you taught how to prioritize between two good activities?
  • Were you taught how to plan a project?
  • Were you taught how to manage the information that would help you make a choice later?
  • Were you taught to take action on your own, based on information you generated?

Or did you develop your self-management knowledge on your own, through experience?  If so, have you thought carefully about how well your practices work? Have you experimented with your practices? Are you actively trying to improve them?

I think self-management is an area in which we can dramatically improve our productivity, if we challenge some long-held assumptions and seriously assess our practices. The current work environment is calling us to do so.

Question: As a knowledge worker, how do you manage these new challenges?  Whether you were trained to do so or developed your own methods, tell us how and where in the comments.

Leave a comment
Related posts
Getting Knowledge Work to Done
May 6, 2017
Time According to Drucker’s Watch
November 19, 2016
6 Reasons Planning Is a Pain (and Why You Should Do It Anyway)
November 9, 2016
Books, Productivity

Deep Work by Cal Newport

qtq80-VjzbQmI have followed Cal Newport for several years now. He earned a Ph.D. in computer science at MIT and is now a tenured professor at Georgetown. He was doing his doctoral work at the same time I was and seemed to have good advice on how to go about earning a degree without going completely insane. He has since published a pair of great books: So Good They Can’t Ignore You and Deep Work: Rules for Focused Success in a Distracted World. I’ll post on So Good later, but my thinking about attention management is closely tied to his notion of deep work, so I wanted to cover the topic and the book for my readers’ reference.

Deep Work explains one way in which work is changing. With the rise of knowledge work, Newport sees a division of labor among knowledge workers. Much of knowledge work is what he calls shallow work. Shallow work is the administrative side of knowledge work, including managing email and other communication, organizing, editing, and non-directed surfing on the internet.

In contrast, deep work is the work knowledge workers produce when they are at their best. This is the demanding problem solving, strategizing, proving-new-things kind of work that produces value. It is valuable in the economy because it is hard to replicate and to the worker because it improves the worker’s skills. Deep work occurs in relatively isolated extended chunks of time. We work to get our heads around the entire problem, chapter, or situation and, when we finally do, make progress on the really difficult part of our work. It is the kind of work that, if interrupted, takes a significant amount of time to restart because we have to regain the mental state that was lost due to the interruption.

Newport argues that, in order to produce value, we need to spend more time in deep work.

The problem is that social and technological forces move us toward shallow work. Three prevailing, but questionable, ideas push us toward shallow work: We can multitask and still produce our best work (we can’t). We have to be available for communication at all times (probably not). Social sites and most of the internet produce valuable information (they don’t usually). Newport acknowledges that exceptions exist, but generally argues that these three ideas are not true. Finally, he notes another force toward shallow work – our brains favor shallow work because it is cognitively easier, requiring less mental effort and concentration.

Newport argues that deep work is valuable, increasingly rare, and more meaningful than shallow work.

The value of deep work is based on two of its properties: it allows us to quickly master hard things and it allows us to produce more output of higher quality. It is increasingly important to do these things as we face cognitive computing and the automation of manual and some craft work.

The rarity of deep work is based on cultural arguments. 1) Workplaces are trending toward open offices, instant messaging use, and increased insistence on social media presence, each of which hampers deep work efforts. 2) Particularly in large organizations, it is difficult to directly measure the contribution of deep work; this supports the cult of busyness, in which visible, but shallow, contributions are often rewarded. 3) The cult of the internet believes that something is irrelevant if it is not visible.

In support of the meaningfulness of deep work, Newport makes three different arguments, backed by research.

  1. Neurology – Managing our attention through engaging in deep work keeps us focused away from the traps (e.g. the comparison trap) that tend to make us discontent.
  2. Psychology – Our minds prefer to be engaged in deep work; we like to be challenged and get bored quickly with routine tasks and what currently passes for leisure.
  3. Philosophy – Deep work promotes a sense of craftsmanship (the technical version is elegance) which endows our work with meaning far beyond simply earning money.

Newport finds that, in order to engage in deep work, we need to be in relatively distraction free environments. He cites many examples of authors, academics, and scientists who produced world-changing work, in part, because they removed themselves from distraction rich environments. In part 2, Newport lays out four rules that help us determine what is deep work in our context and help us focus on it. This is the point where my work connects to his – attention management. Much of part 2 relates to thinking on this site.

Rule 1 covers methods of entering into deep work periods. We need to defend our attention while we are trying to work deeply.

Rule 2 shows ways that we can improve our deep work stamina. We need to train ourselves to focus and concentrate, particularly since if we have been engaging in shallow work for years.

Rule 3 is about our communication tools. We need to be more selective about what we engage with on the internet through recognizing that not all tools are of equal value.

Rule 4 considers removing shallow work. We need to identify shallow work in our lives and work to eliminate it; Newport gives a couple dozen tips and practices to try in our work lives.

The book concludes with a short description of how development and application of deep work principles have fueled Newport’s quite successful academic career to date. That evidence alone should be sufficient motivation to bring some deep work into our lives.

Cal continues to elaborate on these ideas on his blog. In addition, Brett McKay interviewed Cal on the Art of Manliness podcast, here.

Related posts
Getting Knowledge Work to Done
May 6, 2017
Time According to Drucker’s Watch
November 19, 2016
6 Reasons Planning Is a Pain (and Why You Should Do It Anyway)
November 9, 2016