Why teaching burns you out differently

Ask a teacher in mid-November how they're doing, and you'll get an answer that sounds like it's coming from someone completely wiped out — even though, on paper, they've worked maybe twenty-two hours of direct teaching that week. From the outside, that doesn't add up. From the inside, it's completely obvious. A teaching day isn't a shorter version of an office day. It's a different kind of day, with a different structure, a different load, and different ways of getting through it in one piece.

Most productivity advice is written for someone sitting at a desk who can move a meeting and decide when to take a break. A teacher has none of that. They can't push back the start of third period, can't step away from hallway duty, can't ask twenty-five fourteen-year-olds to wait a moment while they finish a thought. Their day is a fixed grid, and everything else has to fit into the gaps — and the gaps are short, loud, and full of people who need something right now.

Teacher burnout isn't the result of hours worked, but of the job draining a resource we don't know how to measure. Hours can be counted. The capacity to attentively respond to twenty-five different people, switch between roles several times before lunch, and simultaneously keep track of who's having a bad day — that can't be counted. And yet it's exactly that capacity that runs out first. This chapter names why that happens and what follows from it for your defense. You'll find concrete tools and routines in the next chapter, on preparing lessons as a system, and in the tips we'll reference along the way.

A day that can't be planned in hour blocks

The basic unit of a teaching day is the lesson — forty-five minutes, followed by a ten-minute break. That grid is enormously practical for running a school and enormously impractical for any kind of focused work outside the classroom. It means the longest uninterrupted block a teacher normally gets is forty-five minutes — and even that is "free" only if there happens to be a gap in the schedule.

Focused work carries overhead. It takes about twenty minutes before you're actually deep in thought about a complex lesson plan or how to rephrase an exam question. Right about when you get there, the bell rings. The economics are unforgiving: out of a forty-five-minute window, roughly fifteen minutes of real work remain. Two such windows back to back yield a full hour of work — more than double, because you only pay the startup cost once. That's exactly why two connected free periods can't be replaced by four scattered ones.

On top of that come inputs that keep chipping away at the grid. Hallway or cafeteria duty. A student who wants to ask about a grade. A colleague who needs to swap a supervision slot. Admin who needs to know by lunch who's going on the field trip. A phone call from a parent. Covering a colleague's lesson you find out about at eight in the morning. None of these things is unreasonable on its own. Together, though, they mean a teacher can practically never be sure the next half hour will go the way they planned it.

And it's not just about lost time. Every interruption has a cost you keep paying after it ends: the brain returns to the original task slowly, and first has to rebuild the context. For work that's pieced together all day out of fragments, that becomes a permanent state of mild fuzziness. The mechanism of context-switching is covered in more depth in The psychology of productivity; in school, its impact is simply sharper, because the interruptions don't come from notifications but from people you cannot turn away.

Emotional labor: twenty-five people who all need something

The second thing specific to teaching is talked about least and costs the most. A teacher doesn't work with data, documents, or projects. They work with people — not one or two, but twenty-five at once, repeatedly, all year, in a role that demands a consistently even manner regardless of how they actually feel.

The literature calls this emotional labor: the need to regulate your own expression to match a role, rather than your actual internal state. A waiter has to smile. A nurse has to stay calm. A teacher has to be fair, patient, and legible even when they're angry, in a hurry, or have a sick kid at home. Burnout research repeatedly shows that this specific kind of strain — a sustained gap between what you feel and what you show — correlates with burnout more strongly than sheer workload does.

And it multiplies by the headcount. A classroom doesn't run one relationship; it runs twenty-five parallel relationships plus the dynamics between them. A teacher is continuously assessing who isn't working the way they usually do, where a conflict is brewing, who needs the instructions repeated differently, who's trying to get attention and why. None of that shows up on a timesheet. All of it consumes attention and decision-making capacity — exactly what's then missing in the afternoon, over lesson prep.

Add role-switching on top. Over the course of one morning, a teacher is a subject-matter expert, a group facilitator, an investigator of a minor conflict, an administrator, an evaluator, and sometimes the only adult who notices that something's wrong with a specific kid. Every switch has its own overhead. By the end of the morning it shows up most in small decisions — you start picking the easier option, not the better one.

Work that's never finished

Most professions have a state called "done." A project gets delivered, an invoice gets issued, a case gets closed. Teaching doesn't have that state. Every lesson you teach could be prepared better. There's always a better opening question, a clearer example, a fairer assignment, more precisely targeted feedback on that specific essay. The line between "good enough" and "could be better" is nowhere fixed — the teacher sets it themselves, every day, for every lesson.

That's especially treacherous for conscientious people. Anyone who takes the work seriously keeps pushing that line higher, until they hit physical exhaustion. And because feedback arrives slowly and ambiguously — you don't immediately know whether better prep actually led to better learning — there's no signal telling you "one more hour of work here won't add anything." The effect ends up being the exact opposite of what you'd expect: the most burned-out teachers are often the best ones.

The practical defense isn't to stop trying, but to set the limit in advance, not by feel. Decide how much time a given type of prep deserves, and do your best work within that time — instead of letting the time run until you personally judge it's good enough. That maps onto sorting prep work by return: a lesson you'll teach once this year and never again deserves a different investment than a lesson you'll repeat for five years across four parallel classes. The next chapter, on systematic lesson prep, develops this logic further.

The same principle applies to grading. "Grade it well" is a bottomless category, because you can always write a longer comment on any piece of work. "Grade it so the student knows exactly what to improve next time" is a category that has a bottom. The first leads to a Sunday evening buried in a stack of notebooks; the second leads to usable feedback in a fraction of the time. Practical methods for creating and grading tests are covered in the tip Tests and quizzes with AI.

A job that never ends

The schedule has an end. The work doesn't. Notebooks get carried home, lesson prep happens in the evening, a parent's email arrives at half past eight on a Sunday, and the school chat pings at ten on a Tuesday night. Formally, no one ever said a teacher is available all the time. Practically, it's expected — and worse, teachers themselves have practically gotten used to treating that as the norm.

An unbounded job has two effects, and both are subtle. The first is obvious: work eats into free time. The second is worse — there's never a moment when your head gets to register that you're done. Unfinished things stay open in the background and consume capacity even when you're not actively working on them. So on a Saturday you're not working, but you're not resting either; the whole day, a list of unfinished things keeps smoldering in the back of your mind.

Communication with parents is especially sensitive here. An email that arrives on Sunday usually doesn't need a Sunday reply — it needs a reply within a reasonable window. The difference between "I'll reply within two working days" and "I answer whenever my phone buzzes" is huge, and the teacher sets it themselves through how they behave in the first three months. Whatever expectations you allow to form in September, you'll be serving in May. For how to set up clear communication with parents without endless explaining, see the tip Grading and parent communication with AI.

The same applies inside the school. When a teacher is always available on chat, availability stops being a signal of urgency and becomes the default state. Colleagues ask right away because it's faster than writing it down. Admin sends questions in the evening because that's when they have them. And because everyone is acting in good faith, no one thinks there might be a problem — until someone states the boundary out loud.

September to June: a curve worth knowing

The teaching year has a shape that repeats and is regular enough to plan around. Ignoring it means planning December just as generously as September and then wondering why it didn't work out.

The practical consequence is simple: demand and capacity don't move together over the year. The most work arrives when there's the least energy for it. Anyone who knows this in advance can shift part of the work forward into September and October — prepare tests ahead of time, build templates, record or write up materials that will come in handy in May. The point isn't to work more, but to move the work to where it's cheaper.

The second consequence concerns burnout. It doesn't arrive suddenly; it arrives as a slow decline in interest and rising cynicism that's easy to keep explaining away as ordinary tiredness for a long time. The warning signs are fairly concrete: reluctance to go to school in the morning that doesn't go away even after a free weekend; irritation at students who aren't doing anything wrong; a fading curiosity about what the lesson will actually accomplish. When these show up, that's information about system load, not personal failure — and it's addressed by adjusting your routine and having a conversation, not by trying harder.

Energy instead of time as the metric

All of this points to a shift that matters enormously for teachers: plan by energy, not by time. In an office, time is a reasonable proxy for capacity, because the load is fairly even. In school, even that doesn't hold. Two free periods after four taught periods with a ninth-grade class are not the same as two free periods in the morning before teaching starts, even though the periods are the same length.

The practical exercise is undemanding. For three weeks, note after each day which part of the day you were actually able to think and when you were just running the operation. A pattern shows up quickly and tends to be surprisingly stable: most teachers have one strong window in the morning and a weaker one after the afternoon break, with the period right after a tough class usable only for mechanical work. Work gets assigned accordingly — demanding prep and thinking go into the strong windows, copying, filing, and routine into the weak ones.

That also leads to a sensible way of handling free periods in the schedule. Scattered free periods are nearly worthless for focused work; two connected ones outperform four scattered ones. When the schedule can be influenced, it's worth actively asking for connected windows and justifying it with the work, not with comfort. When it can't be influenced, it still makes sense to at least sort tasks by what kind of window they need — and not try to cram something that needs an unbroken half hour into a ten-minute break.

Four lines of defense

The defense against teacher burnout doesn't rest on one trick. It rests on several decisions that reinforce each other and that need to be made once and then held.

First line: protect unbroken blocks. A free period isn't a reserve for whatever comes up — it's a work block whose content is decided in advance. The difference between "I have a window, I'll do something" and "I have a window, and in it I'm finishing the lab assignment" is the difference between a lost hour and a used one. The block deserves physical protection too: work somewhere other than the staff room, where the odds of interruption are highest, and keep your phone zipped in your bag. What that kind of setup looks like at home is described in the tip Deep work mode at home.

Second line: an end-of-day ritual. Work with no natural end needs one imposed artificially. Ten minutes, one fixed sequence, the same every day — and then it's genuinely over.

A detailed version of this ritual, along with why it works, is covered in the tip Shutdown ritual.

Third line: boundaries on availability. Set them in advance, state them out loud, and act accordingly. Concretely, that means deciding by when you reply to parent emails (say, within two working days), when you're available to students (a fixed office-hours slot beats "whenever you catch me"), and what counts as a real emergency that justifies a phone call in the evening. A boundary only others know about, that you don't hold yourself, doesn't count; a boundary you hold but never told anyone about reads as unreliability. You need both.

Fourth line: deliberate use of the staff room. The staff room is the most valuable support resource a teacher has — a colleague who knows the same class can save you an hour of thinking with one sentence, and sharing materials in the department office is about the best investment you can make in your own time. At the same time, the staff room is the environment with the highest density of interruptions, and a place where it's easy to spend forty minutes complaining and come out more tired than before. The distinction is simple and worth making consciously: am I going to the staff room to solve something and recharge, or am I sitting there because I just arrived? Sharing works best when it has a form — a shared repository of materials and agreed-upon formats are more concrete support than good intentions. A related topic is covered in the chapter Team productivity.

Where AI helps and where it doesn't

AI won't shrink your class size, won't take duties off your plate, and won't teach your lessons for you. What it can do is take on a substantial share of work that's time-consuming and intellectually unrewarding: a first draft of a lesson plan, variant assignments, rephrasing a text for a struggling student, a proposed project structure, a draft reply to a parent. That is, the work that eats into a teacher's evenings without leaving behind the feeling that you did the part that actually mattered.

One rule applies here that matters more in schools than almost anywhere else: AI proposes, the human approves. Material that goes to students, and text that goes to parents, must pass through the teacher — not just for factual accuracy, but because the teacher is accountable for the content and only they know what that particular class needs. And a second rule that's non-negotiable in a school setting: students' personal data, evaluations, and sensitive family circumstances don't belong in an ordinary tool. Work with anonymized prompts, and if you need to work with real data, only in a paid account with contractual data protection and in line with your school's policies. For a practical approach to using AI for lesson prep, see the tip Lesson prep with AI; the tip Administrative work with AI covers the admin side.

Key takeaways

• A teaching day isn't a shorter office day. The forty-five-minute grid, duties, and unpredictable interruptions mean an unbroken block is rare — and two connected windows outperform four scattered ones.
• Emotional labor takes the most: keeping an even manner toward twenty-five people at once, all year. That load never shows up on a timesheet, but it's the first thing to run out.
• Work that's never finished needs a limit set in advance. Not "until it's good," but "this type of prep gets this much time."
• Set your own boundaries on the job: an end-of-day ritual, a clear turnaround time for parent replies, and a fixed office-hours slot for students. Expectations created in September get served in May.
• Plan by energy, not by time. Demanding thinking goes in your strong windows, routine in the weak ones — and expect the year's curve to dip exactly where demands rise.
• The staff room is both the biggest source of support and the biggest time sink. Go there with intent, not out of momentum.