How I Control Stress at Work

  • Optimise for productivity over predictability, because if you’re not productive, it will be stressful for you.

  • Try to identify, call out and communicate risks in a project as early as possible. Otherwise, the project may be derailed, causing stress.

  • Avoid setting up processes where people or projects have to get your approval before they can move on to the next stage. That adds work for you. And stress — because you can’t have people sitting idle, you’re forced to accommodate them no matter your other commitments. Approval processes also prevent them from learning and so doing more in the future, in turn adding more work for you. Instead, tell people to inform you and proceed, rather than waiting for your approval. Asynchronous course-correction is better than synchronous approval, as a general practice.

  • Increase flexibility, such as by telling your sales team not to pre-commit features to customers. Inflexibility causes stress. If there are two ways of doing something, and one is more flexible, and has no significant downside, go with it. Don’t trade flexibility away unless you get something valuable in return. If you’re choosing between two cars to buy, and one of them is more comfortable, you’d buy that unless the other car has other benefits. You wouldn’t give up the benefit of comfort unless you’re getting something in exchange. Flexibility is the same thing — don’t give it up unless you get something valuable in exchange.

  • Another way to increase flexibility is to ask for meetings to be recorded and watch or listen to them later, at your convenience.

  • Spend money for time or flexibility. For example, I bought an audiobook version of a paper book I already own, so that I can listen to it while doing housework. This effectively creates free time for me.

  • People are often not clear about what they want, or what the consequences will be of what they’re advocating. So when I’m asked to do something, I hear them out, ask clarifying questions, propose different options to try to bring out what they have in mind, and generally explore the territory with them, because people are not often clear a priori. Once we’ve reached a common understanding, I walk people through the consequences of what they’ved asked for. If we’re trying to decide which car to buy, I might say “You said you want to go off-road. That’s an SUV, and it will cost more. It will also have worse mileage. Are you okay with that?” If people don’t accept the tradeoffs of what they’re asking for, there’s no point continuing the discussion, because we’re no longer in the real world; we’re in a dreamworld where there are no tradeoffs. As I’ve told one stakeholder, “You can ask for whatever you want as long as you understand and accept the consequences.”

  • Be transparent, and communicate publicly what you’re planning to deliver. A lot of stress comes from mismatched expectations, so set expectations clearly, and manage them periodically. A CFO said that if you had a good quarter, but people were expecting more, your stock price will go down. Conversely, if you had a bad quarter, but people were expecting worse, your stock price will go up. So it’s not about whether you had a good or bad quarter in absolute, but relative to expectations.

  • Insist on transparency from people you work with, so that you don’t have to worry what they’re thinking.

  • Delegate as much as possible.

  • Limit your meeting time. Meetings make you reactive, and walking in to meetings without having done your homework wastes everyone’s time. So I’ve set up 2 ½ hours of meeting time a day. And blocked the rest in calendar, including lunch. When someone asks for a meeting, I ask them to schedule it any time my calendar says is free. So, automate things this way. It also avoids the awkwardness of saying no again and again, or seeming inflexible as a result.

  • If you need to prepare for a meeting, schedule the meeting after you’ve prepared, not before. That reduces unnecessary pressure on you. For example, I was supposed to give a bunch of tech talks. I gave 4 of them, but then I got busy with other tasks for weeks, which was perfectly fine because they weren’t scheduled ahead of time. I can now prepare at my convenience rather than being bound to a scheduled event.

  • Spend some time investing in your own learning and growth. It can be a small amount of time, such as an hour a week, or a day a quarter. This can be reading a book. Or watching some talks on Youtube. Or discussing your work with a peer and seeing if there’s a different way of doing it. Don’t feel guilty about it — improving yourself is in your company’s interest. The thinking that “we’re a startup, so we don’t have time for that” is wrong. It’s like saying we’re in a hurry to reach Chennai, so we don’t have time to refuel. Too many people in startups are on the treadmill, working inefficiently, generating less value for their company, sometimes zero, sometimes even negative! If they improve their productivity by 10%, taking multiple days off from routine work, that investment will repay itself soon, so it’s in the company’s best interest. Give yourself permission to do this. Nobody else can.

  • Autonomy makes us happier, reducing stress:

  • A sense of purpose or mission makes us happier, again reducing stress.

  • Don’t tolerate negative behaviors like people who tend to blame before getting the facts, make you feel inadequate, don’t support you, criticise you not backed up by facts, are distrustful, non-transparent, political, etc.

  • Before you take up a job, get a feel for whether the people in the company are relaxed or stressed. If they feel like they’re under immense pressure, pushing a boulder uphill, their stress will affect you, so don’t join that company.

  • Before taking up a job, check if their working styles, attitudes and expectations align with you, because misalignment causes stress.

Prioritisation

  • Be cognisant of the tension between the time things take and when you need them done. People often decide they need something done in a week, so they give themselves a deadline for a week, but if it takes two weeks, the deadline causes stress. If you need something done in a week, that doesn’t make it doable in a week. Things take the time they need irrespective of what you need. If you go to a showroom to buy a car, but have only $1000, arguing with the salespeople and telling the car company that they need to meet your budget won’t work. Likewise, your need to get something done in a week doesn’t change the reality. If you try, you’re only stressing yourself, which is counter-productive. And it will ultimately take two weeks anyway, in which case you’re better off not being under the illusion that it will take a week. A lot of stress in startups comes from trying to dictate terms to reality. Instead, accept it, like you accept the delay caused by traffic, rather than trying to control it.

  • If someone tells me they need me to do something in a fortnight, and I feel it’s unrealistic, I ask them whether they want to reduce scope (such as number of features) or quality (UX, bugs). If they’re not open to these tradeoffs, I tell them it will take what it takes and stop the discussion.

  • Things always take longer than expected, so keep that in mind when planning and setting expectations both to yourself and to others. A rule of thumb is to take your estimate and double it.

  • Before taking on a project, think twice about whether it can fail. For example, if you have a notes app, and you want to add attachments to it, there’s no reason it will fail. It can take more time than you expect, but there’s no reason this effort will fail. So it’s safe to take on. But many projects are not like this — they have hidden risk. So be cautious. Think twice before taking on any task that can fail, because it will stress you. You have to deliver results, not just put in effort, and when things take effort but don’t yield results, it puts extra pressure on you. You can still take on an effort that can fail if you’re conscious about it, and treat it as an experiment. Time-box it by allocating a fixed amount of time, maybe a day, to it. When it’s over, you stop. Be clear to yourself that it’s okay for an experiment to not produce any results, because that’s what an experiment is. Call it an experiment in your discussions with other stakeholders, to set expectations. There are two factors in deciding to take on a project: how long it takes and how likely it is to fail. Take both into account: a big project that won’t fail is fine, and a small time-boxed project that can fail is fine, but think twice before taking on a big project that can fail.

  • Be cognisant of your emotions. If you feel panicky or rushed or impatient, it won’t make the work get done any faster. All it does is add tension and unhappiness, making your month bad. You should work calmly and methodically, like a scientist doing an experiment, without counterproductive emotions like that of a deer running away from a lion.

  • Avoid giving deadlines, whether to other people or to yourself. Deadlines cause stress because you may not meet them, for two reasons: first, the task may require more time, as we saw above. Second, new tasks may come up that are more important. In that case, you have a problem: you either break your commitment (thus affecting people who planned based on your deadline) or work on the original, less important task because you’ve committed to it, neither of which is a good choice. You want to give yourself the flexibility to work on what is important today, rather than what you guessed a week back (when you had incomplete information) will be important today.

  • If you must give a deadline, add a big margin of safety. If something takes two days, I’ll say that I’ll do it in a week. That way, even if it gets delayed, it will be done on time.

  • When you give an estimate, make it clear it’s only a guess.

  • Some deadlines are fixed. For example, if you need to plan tasks for a sprint starting on Monday, be ready the previous Monday. Don’t put it off till the last minute, because things will invariably go wrong. So have buffers everywhere. If you need to come up with 20 tasks for the next sprint, come up with 30, because some tasks will get bogged down. Don’t try to pack things in tightly, because one will go wrong, causing the whole thing to fall apart like a bunch of dominoes, causing stress. Instead, keep things flexible and decoupled so that you can accommodate reality without having things thrown into disarray.

  • Prioritise saving clock time over calendar time. What does that mean? Suppose you have a task that takes 2 hours, but has a dependency on something else that will be ready in a fortnight. So you can wait till then to do it. Or you can do it now, spending 3 hours to do it in a different way. What should you do? Unless the task is of the topmost urgency or importance — and the vast, vast majority aren’t, by definition — wait a fortnight. You can do other things in that fortnight. Whereas if you waste an hour today, you will never get it back. Save clock time in preference to calendar time, as a general rule.

  • Don’t confuse urgent for important. Urgent is how soon it needs to be done. Importance is how much you’d regret it looking back after an year if you didn’t do it. An example of an important but not urgent task is getting to know your team as people outside work, because that builds connection and trust, which defuses disputes. Don’t confuse the two, or let the urgent override the important, as many of us do, including in our personal lives. Deadlines are a tool to rank tasks by urgency, and prioritisation is a tool to rank tasks by importance. People often misuse deadlines, applying them for all tasks, thus overweighting one aspect — urgency. To guard against this, whenever someone says “This task needs to be done in a week”, I’m automatically skeptical, because people often have a false sense of urgency.

  • One source of stress is trying to be reactive and taking on work as it comes in, which results in doing too many at once, causing stress. Instead, be proactive: maintain a list of priorities in a Google sheet. When a new task comes in, instead of taking it on right away in addition to everything else you’re doing, which overloads and therefore stresses you, let it go into into your list, and wait its turn. Pick up a new task as a conscious decision, not automatically like a call center operator answering the phone when it rings.

  • Maintaining your list in a spreadsheet lets you offload the tasks from your brain, where they cause stress if they linger. You can’t remember it all. You have limited mental capacity.

  • Prioritising in a spreadsheet also helps you be more methodical, and recognise multiple aspects of each task. Such as urgent vs important. And whether you love doing it. Which should have some weight, not zero. We’re not emotionless robots. Our morale matters to our productivity. And to enjoying our journey, because it’s not just the destination that matters. So, putting down your tasks in a spreadsheets helps you make more methodical decisions to tackle them in an orderly way, as opposed to chaotically.

  • After you’ved prioritised and reached agreement with stakeholders, say no to stakeholders who speak up partway through and ask for something else in addition to or instead of what was agreed upon. If the situation changed, that’s reasonable, but otherwise, it’s not — if you keep changing focus, you won’t achieve anything, and you’ll be held at least partially accountable for the resulting unproductivity. So say no.

  • If you’re asked to do more than you can, say no and push back against people. Sometimes I say no. Sometimes it’s a more nuanced no like “not for the next few months”. It’s better to ruffle a feather than to stress yourself, or make people happy now but not deliver later.

  • When you say no, try to offer a choice. For example, if I can do only two tasks in the next month, but I’m asked to do four, I tell the other stakeholders to pick two, or pick two myself but tell them I’m flexible about which two will be picked, as long as we recognise that only two can be done. This comes across as less autocratic than saying “I’ll do A and B, not C and D, and you have no say in this.” You do want to take everyone’s input into account, and you don’t want to make stakeholders feel as if they have no say, at an emotional level.

  • If someone disagrees with your decision, ask them to put an alternative on the table for consideration. That puts the onus on them. They often realise that they don’t have a better solution, defusing the disagreement. And if they do get back to you with a proposal, point out its pros and cons and ask if they’re okay with the cons. If so, you can consider it. This also creates a culture of proposing alternatives rather than complaining (even if well-intentioned).

  • If someone wants to overrule you, insist that they take responsibility for the decision. If people disagree vehemently in a meeting, pick one of them and ask them if they want to select one of the options on the table and accept responsibility for its consequences.

  • Don’t be emotionally attached to decisions. Say what you think is best, and why, but if a different decision is made, and people understand and accept its consequences, let go. We’re not always right. And sometimes it’s a question of a tradeoff more than right or wrong, and reasonable people can disagree.

Work-life Balance

  • Take all the leaves you’re entitled to. If your company has a policy on how many leaves you can carry over to the next year, make sure you don’t lose any leaves by exceeding this number. I’m ending 2020 with a vacation balance of zero. Be as serious about your leaves as you’re about work.

  • Prioritise work and life equally, not one above another. Your life is not something that to be done in the time remaining after work.

  • Work 40 hours a week. If you work more (or less) in a given week, make up later.

  • Be an employee rather than a founder, consultant or owner of an agency. Have a predictable salary coming in every month.

Limit Access To Work Accounts

  • Try to log in on as few devices as possible. For example, I have two computers, and I picked one to use for work. I never log in to my work accounts on the other one.

  • Don’t log in on your phone unless you actually work from your phone, like attending meetings when you go for a walk. Wanting to check Slack doesn’t count as work.

  • Log out when you decide not to work, certainly when you go on vacation, on Friday evening and at the end of the day. If you can’t log out on all your devices, log out on the one you can.

  • Make it easy to log out of all work services at once. I use Edge for this purpose, and clearing all cookies logs me out of Google, Slack, Mixpanel and more. Using a second browser makes it easy to out of work while remaining logged in to my personal accounts in my primary browser. Remember that the goal is to make it as easy as possible to log out. Because if something is hard to do, you won’t end up doing it — you’ll take the path of least resistance.

  • Make it hard to login, such as by enabling 2-factor authentication. Friction is good! That may make you think whether you really need to log in, rather than reflexively logging in to check if there are any updates.

  • If you need to log in, log in only to the services needed.

  • Don’t remain logged in to Slack all day, because you don’t need to be available every hour of the day — that takes away focus from your work. Remember that you’re paid to get work done, not to chat on Slack. Communication is only a means to an end, not the end. My solution is to be intentional about opening Slack. I have to have a reason for opening it. If I feel like opening Slack, I stop and ask myself why I want to open it. For example, I may need to send a message to someone, so instead of remembering it in my mind, it’s better to get it out. So be intentional about opening Slack. And once I’m done with Slack, my rule is that I have to log out. This increases friction to logging in again, which is exactly what you want. Seamless is bad! Companies like Slack are incentivised to boost their engagement numbers, not act in your best interest. We’re all addicted to coordination, which is bad for the same reason being addicted to alcohol is bad. I’ve told my peers that I won’t log in to Slack every day, and I’ll respond tomorrow, but I won’t forget, and that can always call me if something is urgent. Set such expectations so that everyone is comfortable with you logging out.

  • Don’t use personal accounts for work. Because that makes it hard to disconnect from work.

  • Don’t have an email app on your phone. That frees you from work email, and from personal email, too, on the go.