Deadlines Are Killing Us, And Almost Everything Else I Know About Leadership

This morning, I spoke with a good friend who lives in France. He started a business with his wife to develop a web-based, software-as-a-service product. As the business began to grow, his belief in it increased, and he started to take it more seriously. One of the things he did next was to start planning and monitoring progress. To do that, he started setting deadlines. But soon “missing deadlines” created an enormous amount of stress for both him and his wife, and for the other people they were working with. “It’s just not worth it,” he told me, and he backed-off from the deadlines.

Before he and I talked about this topic, my friend had been thinking that he needed to figure out how to use deadlines, to discover how to “make them work.” After our conversation, and once he had digested the conclusions we had come to, he told me by text message that “deadlines simply don’t work at all.” Now he plans to print a massive sign that reads NO DEADLINES and stick it on the wall of his business.

In this article I’m going to explain to you what deadlines really are, why they are not effective, and what the alternative is. Actually it’s not really an alternative because what we want is a way to achieve world-class creativity and productivity, not simply an alternative way to thwart it, which is what deadlines do. But first we need to take a look at what work is and what motivates humans to do it.

In any endeavor, progress is made through the completion of a sequence of small actions, actions ultimately instigated by humans. Any completed project is constructed from a framework of these actions, stacked on top of each other, constructing a solution that is far more valuable than the sum of the values of each of those small actions considered individually.

Even though many operations can now be automated, progress is still dependent on a creative agent initiating, monitoring, fixing, and checking the results of many automated processes. Even as increasingly sophisticated machine intelligences automate aspects of work that were historically performed by humans, we must still be the ultimate initiators of those processes. Until we have perfect slave intelligences, work will essentially be completed only at the whim of meaty animals containing large neocortices whose actions are strongly influenced by ancient reptilian and mammalian brain structures that are vastly more interested in eating things that move and fucking things that breathe than on manipulating the numbers in an Excel spreadsheet.

By the way, I don’t think we’ll ever have artificial general intelligences (AGIs) that can replace humans without treating those AGIs as autonomous beings (and paying them). As I wrote in ¹ (link at end), I believe that intelligence in any form is synonymous with freedom.

So how do we get people to do things? Well, countless psychologists have studied this topic and the results are pretty conclusive; and they seem to be intuitively correct in hindsight. As individual workers, people need (1) to have autonomy, (2) to feel that they are on a path to some kind of mastery, and (3) to also find meaning in their work. This is explained extremely well in a book called ². These three factors are the bedrock of motivation, and I believe that as we create machine intelligences that meet and exceed the capabilities of humans, our ability to provide these environmental factors to our super-intelligent robot progeny will be crucial.

But people don’t operate on their own. We curate teams of people with diverse skills and perspectives, creating many-brained flesh organisms that can achieve outcomes that a single biped couldn’t. We also have hierarchies of people, as represented by org charts, that capture increasingly broad domains of responsibility and authority.

So how do we get people to do things in a group context? Well, the scientific evidence is pretty clear on this too. First of all, we don’t hire assholes. If we do accidentally hire an asshole, we identify and fire them as rapidly as possible. Finally, if by some stroke of very bad luck an asshole does stick around, then no matter how “brilliant” we might think they are, we certainly don’t promote them up the leadership chain.

What is an asshole though? In this context, it’s basically a narcissist, someone who consistently puts their own personal, short-sighted needs in front of the other humans around them and in front of the best interests of the organizations that they are a part of, which includes the employer that pays their salary or wage.

When I write about narcissists below, I’m referring to people with various shades of narcissistic personality disorder (NPD) or similar clusters of change-resistant personality features. Some people exhibit features of NPD without being diagnosable, a state which is referred to as “sub-clinical;” those features are still usually very destructive to relationships. The American Psychiatric Association’s (DSM-5) lists the following symptoms:

  1. Grandiosity with expectations of superior treatment from other people
  2. Fixation on fantasies of power, success, intelligence, attractiveness, etc.
  3. Self-perception of being unique, superior, and associated with high-status people and institutions
  4. Need for continual admiration from others
  5. Sense of entitlement to special treatment and to obedience from others
  6. Exploitation of others to achieve personal gain
  7. Unwillingness to empathize with the feelings, wishes, and needs of other people
  8. Intense envy of others, and the belief that others are equally envious of them
  9. Pompous and arrogant demeanor

People with NPD usually exhibit at least some of these symptoms in a way that is out-of-whack with their real-life qualities or accomplishments. Of course, it’s important not to read this list and start diagnosing yourself or others, but it’s good to be aware of this list in order to recognize when these kinds of behaviors seem to be appearing. There are many other terms related to NPD that come from different psychological schools attempting to categorize the phenomenon, such as malignant narcissist, oblivious narcissist, and covert narcissist, all of which hopefully speak for themselves.

All personality disorders are essentially characterized as being ego-syntonic, which means that the person with the disorder consistently perceives their dysfunctional behaviors as normal and acceptable, even if they also attempt to hide them from selected others (such as those with immediate power over them). This characteristic of personality disorders, especially NPD, make them very resistant to treatment. NPD is particularly heavily armored against change because the idea of needing to change is in itself an affront to the narcissist. The narcissist believes that everyone else needs to change, but not them.

Narcissists are highly effective at self-sabotage in the long-run simply because they cannot understand that what’s best for the group is usually also what’s best for themselves. Doing what’s best for the company is obviously the most effective long-term path to career success for someone who can peer just a little bit past their own fear, jealousy, anger, righteousness, and greed. The only kind of employee you need to instruct to “do what’s right for the company” is either severely disempowered (probably by a narcissist) or is a narcissist themselves. By definition, the narcissist will never do what’s right for the company unless it happens to coincide with their own short-sighted and selfish desires. The solution in either case is to locate the narcissist and terminate their employment.

And it’s pretty easy to root-out narcissists in an organization. For the individual contributor, a thorough review by peers is sufficient. Since NPD is characterized by dysfunctional relationships, the working peers of a narcissist will usually be able to convey the experience of chronic demoralization, upset, and manipulation that they have experienced at the hands of the narcissist. They usually want to get away from the narcissist, and often do so by leaving the company. This can result in great cost to both the company and the victim. I believe that this is partly the explanation for the somewhat surprising discovery by Google that the most significant predictor for the effectiveness of an employee is the quality of their personal relationships outside of work. Larry Page, one of the two founders of Google, told this to a friend of mine.

Narcissistic leaders (including managers and supervisors) may be even easier to root out: all you have to do is ask the people they lead. The “lower” part of what is often considered to be a cumbersome 360-degree review process is usually enough. Since the role of a leader is primarily to lead, all you need to do is asses the effect they have on the people they lead, those directly “below” them in the organization. These are the people who are more likely to actually be getting work done, the people who are closer to the reality of the technology, the market, and the customers. They are one step closer to where the rubber meets the road. If you can get these people to talk about it, which can be challenging, the strongest indicator that they may reveal is fear. Those led by a narcissist are often terrified of the narcissist, of their manipulation and of their narcissistic rage. Later in this article, I will tell you about the hallmarks of effective leadership; with the narcissistic leader, those characteristics will be wholly absent.

It’s important to understand that narcissists are usually very skillful at managing up; they’re often really good at kissing ass. So those “above” them in the organization often perceive them as compliant and pleasant and their organizations as effective. Since it’s almost impossible to create organizations with control groups, the narcissistic leader’s own leader has no effective way of personally assessing how much performance is being left on the table due to those being led by the narcissist feeling terrified and demoralized.

Simply assessing leaders primarily on the effect they have on the people they lead is very likely to produce a massive increase in their effectiveness as leaders, which will have an enormously leveraged effect on the effectiveness of the organization as a whole. Even if you’re not trying to root-out narcissists, the important and effective assessment of leadership will naturally do the job for you. For the non-narcissists, perhaps for those existing gingerly on the autistic spectrum, an effective feedback loop will likely lead to an increase in critical learnable personal characteristics such as social and emotional intelligence.

And it’s not even necessary to have a formal, cumbersome, and likely ineffective review process. At the company where I work, I regularly meet with people at all levels of seniority. From having hundreds of candid conversations with those who are led by others, I have a very clear picture of the relative strengths and opportunities-for-growth of many of the leaders across the company. If you want to read more about assholes, I recommend an awesome series of articles titled ³.

Before I go on to talk about leadership, I want to point out that, for a “good-enough” leader, it takes an enormous amount of courage to be willing to even discover and acknowledge the level of dysfunction about which I’ve written above. It then takes even more courage to confront it. It can be really scary to confront narcissists, because most people can sense the rage hidden just under the surface. It also takes courage to transition people out of an organization. What’s more, if your company culture is conflict-avoidant, and therefore probably filled with conflict-avoidant leaders, then it’s susceptible to infiltration by narcissists, and eventually infestation with them. In these kinds of organizations, red flags will be ignored, minimized, and rationalized away.

One way that narcissists flourish is when a leader claims that “he’s a top-performer, so there’s nothing I can do”. It doesn’t matter how much of a rockstar, key player, or top contributor you think the narcissist is, you can be guaranteed that they have a net negative effect on the organization, and ignoring their behavior will not make it go away. For every unit of goodness that they apparently add, they will subtract at least two from those around them, and probably much more. In fact, this is one of the ways they get to shine: by puffing themselves up while tearing down their peers or reports.

Narcissistic leaders also take credit for the work of those they lead while at the same time dismissing their contributions. As each new discovery is made, it is immediately assimilated into the knowledge-base of the narcissist, as if it magically appeared in their brain overnight. There’s no need to recognize the effort of the folks who are actually doing the work because the narcissistic leader apparently already knew the results that were painstakingly obtained. Initially, this can be confusing to employees because the leader seemed to have asked for work do be done unnecessarily.

From the perspective of a narcissist, the purpose of those reporting to them is only to make them look better, to serve their narcissistic needs. This is one example of what I mean by short-sighted selfishness. The narcissist is usually so lacking in empathy and self-awareness, so consumed by the deep psychological processes driving the narcissism, that they don’t even realize that their behavior will undermine their ability to achieve their personal goals in the longer term. Talented people learn to hide their innovations or to reduce their creativity or productivity, or they simply leave the company. This is anti-leadership.

So you can also bet that one unit of goodness you think the narcissist is adding is in fact only a fraction of a unit. As the results of Google’s ⁴ show, “what makes a team effective at Google” is a set of characteristics that simply don’t mesh with narcissism: psychological safety, dependability, structure & clarity, meaning, and impact. The narcissistic employee might be reasonably effective working alone on a small project that doesn’t involve interacting with other people, and that might be an alternative to firing them, so long as you can keep them quarantined; good luck with that.

The long-term costs of the kind of draining and insidiously manipulative behavior that these narcissists act-out are almost incalculable in their scope and depth. As an organization, how do you even account for the bad karma of inflicting psychological torture on innocent people who simply want to come to work, do a great job, and feed their families? As a leadership consultant, the witnessing of this stuff has led to me feeling like I needed to literally vomit from disgust.

I myself know what it’s like to fire people: as a manager, I have let people go. In fact, whereas it was normal for firing to be done by people at the senior vice president level (my manager), I insisted on doing it myself. “It’s my decision, so I’ll do it” I said. And yet, as an empath, being the instrument of that much distress in someone else’s life wore me down.

As I’ve matured, I’ve come to believe that pretty much any employee, with the exception of a narcissist, can be made productive. The trick to avoid having to fire these people is to simply not hire them in the first place. This is why, even though I’m a senior engineer, when interviewing candidates I’m focussed wholly on assessing their personality. I’m able to do this relatively effectively because I have a Ph.D. in clinical psychology. I also have a lot of personal experience with narcissists.

So then what does good leadership look like? The research shows that the most important leadership behaviors are those that nurture increased employee engagement. “Engagement” in this context is part of a very specific psychological term; it doesn’t mean “happiness.” According to Wikipedia, employee engagement is “a fundamental concept in the effort to understand and describe, both qualitatively and quantitatively, the nature of the relationship between an organization and its employees.” Therefore, any leader’s primary objective is the curation of that relationship. Wikipedia’s excellent definition continues:

Towers Watson conducted a thorough and high-quality study on ⁵. In alignment with the prevailing consensus, they found that companies with high employee engagement produced significantly higher returns to shareholders than average companies.

This study also concluded that employee engagement is driven by the quality of the relationship between supervisors and those they immediately supervise, which is characterized by the level of:

  • My immediate supervisor understands and assesses my contributions fairly.
  • : My immediate supervisor communicates openly with me and encourages me to make suggestions, which she takes seriously.
  • : I trust my immediate supervisor, and she trusts my judgement.

I highly recommend reading the full, two-part study. It’s eye-opening and applicable to boosting the bottom-line of any organization.

With all of this information, we now understand where individual motivation comes from in the context of the work itself in the context of the most important relationship at work, the relationship between the supervisor and each of the people she supervises.

This article is intended to be about deadlines, but I felt that it was necessary to clearly define both work human motivation in the context of work before getting to deadlines themselves. This is because deadlines, in our common understanding, are related to people doing work. There’s also no point in rearranging the deckchairs on the titanic if you’re either not steering around the iceberg or patching the rip in the hull. Do the hard part, fix the motivation issues, and you’ll have done 95% of the work.

So what are deadlines? A deadline usually specifies that some measurable outcome will be achieved by a particular date. Without thinking about them too deeply, it’s usually assumed that they’re a way of making a chosen outcome happen by that date. However, when considered more deeply, they can be seen as a way of trying to control human resources. A deadline says, “this is a commitment to provide a very clearly defined set of deliverables by a specific date,” a form of contract. When used in the relationship between a manager and the person he manages, deadlines, being so transactional, tend to rapidly erode all of the qualities that lead to intrinsic motivation.

Unless work is purely mechanical, it involves creativity and requires inspiration, ideally prolonged states of flow. With a sufficient supply of autonomy, mastery, and meaning in an interpersonal context of inclusiveness, communication, and trust, an individual contributor will perform optimally in complex and creative tasks by staying in flow. By the way, any tasks that do not require creativity should be automated-away, applying skill to be rid of drudgery; this should be prioritized.

The primary goal of an effective leader is to ensure that the people they lead remain in flow as much as possible. Deadlines create an artificial constraint that distracts from the true motivation factors, reducing flow, and thereby leading to sub-optimal outcomes in the long-run.

For most of my engineering career, I have enjoyed profound autonomy, and been prolific, while experiencing almost no interference from managers attempting to control my output. I have also been minimally impacted by deadlines. One personal datapoint that enables me to more deeply empathize with those being anti-managed was a short period of being really badly managed. I was reporting to a new manager who seemed to have read a book on management and then decided to do exactly the opposite of everything that was recommended. After I informed him that his daily check-ins were distracting, he switched to setting micro-deadlines for me.

After this manager had left me alone for two weeks with one of these micro-deadlines, we met and I explained to him how much more progress I had been able to make without his interference. He told me that he didn’t agree, because I had apparently not precisely met the arbitrary micro-deadline that he had set for me. If he had been listening to what I told him, he would have understood that, without the distraction of his continual interference, I had achieved far more than his micro-deadline demanded. As always happens with deadlines, the complexity of the unfolding process revealed much more about what needed to be done than could have been anticipated at the beginning, even by me let alone by him.

It didn’t seem to matter to him that I had proactively focused on the most important work during that two week period. What seemed to matter to him was that I had not continued to mechanically focus on the less important work demanded by his deadline. To him, being in control seemed to be more important that the quality, quantity, and relevance of the outcome.

I then happened to complete that micro-deadline a day or two later than planned simply because the reality of the work happened to take a day or two longer than he had originally estimated. His response to my claim that I worked more effectively, and with much less stress, when he backed-off a little on the micro-management was that, “I don’t see the data for that,” apparently having no clue that the words coming out of my mouth were the most valuable data he could ever receive.

The increase in my productivity and the decrease in my stress during those two weeks was due solely to him leaving me alone. Obviously a micro-deadline was just another manifestation of the worst-form of micro-management. Even though he didn’t seem to be interested in gathering the data from me, I tried to convey to him the effect that the micro-deadline had had on me. I remember as the two weeks unfolded that I discovered more details about specifically what needed to be done and the optimal sequence of tasks. As I executed spontaneous micro-pivots, my work, guided by my initiative and autonomy, began to diverge from what he specified in the micro-deadline. I was increasingly focusing on more important tasks that fed into the meta-outcome, tasks that would lead to it being achieved sooner and with higher quality. As the divergence between the reality of the work and his arbitrarily and artificially-generated micro-deadline increased, it caused mounting stress for me. I both had to serve the best interests of the company, driven by my internal leadership, satisfy the pointless and arbitrary specifications of my “manager.” The micro-deadline did nothing but zap my energy, focus, and enthusiasm, and detract from the good work I was doing in spite of his micro-management.

This kind of micro-management might be somewhat effective if you were managing people doing repetitive and non-creative tasks, such as assembling widgets on a production line, jobs that can be, and very rapidly will be, automated-away. This is management which looks like operating a machine, but a machine with human parts. Of course, it’s the kind of management style that would appeal to an unreflective engineer who is used to having total control of his computer, and who is not interested in learning new skills.

This was clearly a low-inclusivity, low-communication, and low-trust relationship, in spite of my repeated attempts to address and fix the issues, to manage up. As a high-performing employee, I found it deeply disturbing that someone who was being paid to supposedly support me was actually doing everything in his power to thwart the only thing I wanted to do, which was to deliver maximum value for the company. After trying to get on the same page with him multiple times, and after discovering that he was resolutely unwilling to change, I left. I was simply not willing to continue allowing him to manage me at the expense of my sanity and the success of the company. I loved what I had been working on, and I loved my team. It was a great loss to both me and to the company.

The experience had been like living inside a room with half-mirrored glass walls. I could see and hear him, but he could not apparently see or hear me. He barked commands at me over an intercom but had no idea what I was actually doing. Nevertheless, he seemed to be vaguely dissatisfied with his own manufactured stories about what I was doing, while also being in control of my air and food supplies: my employment, my salary, my stock grants, and my promotion prospects. Other quality engineers from that group either left or were somehow able to remain in spite of naturally under-performing while suffering in varying states of dissatisfaction and resentment.

Putting my leadership consultant hat on for a moment: in situations like this, I would strongly recommend immediately removing a person like this from management. He didn’t even seem to want to be a manager. He was not interested in people, not interested in their experiences, and not interested in learning about how to lead people effectively. Perhaps he liked the fantasy of power and authority or increased influence that management might bring. Those are not good reasons to become a manager because they are only potential side-effects; they are not the purpose of management.

For someone with my depth of psychological understanding, it’s astounding that nothing was done about a situation like that; he was even given more people to manage! Perhaps there was no leader willing to step up and make the difficult choice. This is another unglamorous aspect of effective leadership: you have to make very hard and often unpopular decisions. Perhaps there was fear of him leaving, even though it would represent a net gain for the company, and a net loss for whatever company received him.

It’s also possible that he might have been a relatively good engineer, although I was never able to determine this because he seemed to always carefully conceal the edges of his knowledge and skills, preventing me from ever being able to assess him properly. It’s possible that he could have been an effective individual contributor.

As well as increasing my empathy for the badly-managed, this experience also gave me insight into what often drives leaders to set deadlines: fear. Emotionally immature and inexperienced leaders, when given the responsibility for the quality and quantity of output from a team, will go straight to their default strategy, which is to try to get immediate control. The deadline is a tool that seems to provide control, but all it actually does is reduce the leader’s anxiety temporarily at the expense of reducing all of the productivity-enhancing human factors in the longer-term. It basically transfers anxiety from the leader to those they lead.

Even though it thwarts the goals of the organization, it feels good in-the-moment for the manager to work with deadlines. But with humans, the more you try to control them the less control you actually have. When we try to control people, they become less motivated, less inspired, and less innovative; they do the bare minimum work or they just leave. So one of the key skills of a leader is self-awareness, to become aware of their own anxiety and to learn to “hold it” and to not act it out by taking immediately gratifying yet self-defeating actions, actions such as requesting, or even setting, deadlines. A leader is effective to the degree that they can resist passing their anxiety to the people they lead.

Learning to lead can be particularly challenging for individual contributors because they are often familiar with managing mechanized processes over which they have almost perfect direct control. In contrast, leadership of humans is like gardening. You plant the seeds and you provide the water, the sun, and the food. The plant itself is the only part of the whole system that can do the actual growing. Giant oak trees grow from acorns, but nobody makes that happen; the acorn does the growing all by itself, but only if the conditions are right. Ineffective leadership looks like a gardener pulling on the little shoots, ripping them up by the roots in an attempt to make them grow taller.

But what we call “deadlines” do actually have some utility. They are a form of contract that can enable multiple organizations to synchronize their efforts, organizations that might be in the same company or in different companies. When you know when my team will deliver product X then you can plan for your delivery of product Y that depends on it. All of this enables the appropriate acquisition and deployment of resources, and it sets expectations for customers. These agreements can be powerful tools if they are used to help people and organizations, but they can easily become inadvertent weapons that harm people and organizations. Let’s extract the goodness from them.

Luckily, the agile software movement has already done this work for us. The book describes an agile contract in which, for a specified price, a particular number of units of goodness will be delivered to a customer by a given date. This enables the development team and the customer to continually adapt to the unfolding technical discoveries of the project, and to the changing desires of the customer; the desires of the customer, whether that’s your own marketing department or another company, will change as the increasingly high-fidelity renditions of the final product are exposed by the scrum process. The result is the delivery of the most relevant, complete, and highest-quality product available in the shortest amount of time and for lowest cost.

When we have a team of humans creating an artifact, we can obviously utilize the motivation tools I describe above. We then need a way for the people to adaptively choose what to work on so that each thing they do maximally increases the value of the product. This is achieved by the scrum methodology of continually organizing tasks in the “backlog” column and then allowing individuals to receive the rewarding satisfaction of taking these clearly important items into the “doing” column before then retiring them to the “done” column.

None of this is based on deadlines. Work takes as long as it takes, and we do everything we can to make sure that the work that is being done is prioritized effectively and that those doing the work are maximally motivated. It’s motivating to know that you’re all working together to add features and improvements to the product that you’ve decided are the highest value to the customer. It’s motivating to know that you are empowered, as a team, to continually pivot, to adjust priorities, and to optimize your ability to maximize customer value.

This is in stark contract with old-fashioned waterfall models of development with fixed deadlines and milestones, where the reality of product development is stifled by being shoe-horned into an almost universally ill-conceived up-front plan. I’ve worked on many projects in which engineers soldiered on, somehow mustering creativity and productivity in the face of depressing slipping deadlines and unavoidable disparities between what was being built and what had originally been specified.

The beauty of scrum is that very early on you have a deliverable product, what is often called a minimum viable product (MVP). Then each piece of work improves that product a little bit. This enables the team to stop at any point and ship the product in the best state it’s ever been in. Some teams using agile development, such as the people making the Facebook mobile app, have a regular release cadence, say one a month, at which point they snap and release a finalized collection of changes. This means that “the train leaves the station” at regular intervals, but it’s not completely decided up-front what exactly will be on that train. The engineers can then focus on adding value to the product knowing that the value will be shipped when it’s ready, which will be very soon. Features are added with hidden on/off switches, so that they can be integrated into the product that will be shipped but switched off at the last moment if a problem is found in final integration testing. They can then hopefully go into the next release.

Agile contracts, whether negotiated one-off project contracts, or regular-cadence release schedules, extract the value of deadlines but leave behind the toxicity. They enable a renegotiable agreement to be made, an agreement that can account for the unpredictable and unfolding reality that is always a part of creative and innovative work. Agile contracts not only eliminate the motivation-squashing aspects of deadlines, they also amplify the motivation-enhancing human factors such as autonomy, mastery, and meaning.


  1. The results of Google’s Project Aristotle
  2. ‘Turbocharging’ Employee Engagement: The Power of Recognition From Managers (Part 1, Part 2) by Towers Watson.
  3. by Jeff Sutherland and J.J. Sutherland

An engineer-psychologist focused on machine intelligence. I write from my own experience to support others in living more fulfilling lives |

Get the Medium app

A button that says 'Download on the App Store', and if clicked it will lead you to the iOS App store
A button that says 'Get it on, Google Play', and if clicked it will lead you to the Google Play store