Teacher workload: What teachers told us they actually need

Teacher workload is the issue that almost every other issue in education quietly sits on top of. Retention, recruitment, wellbeing, the steady drip of departures into other sectors. All of it is downstream of the fact that the working week for a UK secondary teacher is genuinely long and has been getting harder for the better part of fifteen years.

The data is fairly clear. The DfE Teacher Workload Survey has consistently shown UK teachers working well in excess of their contracted hours, with classroom teachers averaging around fifty hours a week and middle and senior leaders considerably more. Education Support, the charity that runs the wellbeing index, has tracked rising stress and burnout year after year. The headline figure that around one in three teachers, roughly 32.5 per cent, leaves the profession within five years of qualifying is a rolling concern in the DfE retention data.

This piece pulls together what the evidence base says about workload, what tends to drive it, and what can change. Some of the levers sit with the individual teacher, but most of them sit at the school or system level.

Three pieces of evidence are worth holding in mind. Together they describe the landscape fairly well.

The DfE Teacher Workload Survey, last conducted in detail in 2019, found classroom teachers in secondary working an average of around forty-nine hours a week during term time. A separate UCL/Allen analysis of teacher hours found around a quarter of teachers reporting they worked more than sixty hours. Marking, planning, and administration were named as the largest non-teaching components.

Education Support's Teacher Wellbeing Index, published annually, tracks teachers reporting symptoms of burnout, anxiety, and stress. The most recent editions have shown around seventy-eight per cent of education professionals describing themselves as stressed, and around 50 per cent saying their organisation's culture was negatively affecting their mental health (Education Support Teacher Wellbeing Index 2024). Wellbeing is not improving in the way the workload data sometimes suggests.

The EEF's evidence on which classroom practices give the best return on time invested is useful here too. The headline finding is that quality of planning matters far more than quantity, and detailed marking has poor returns compared to in-lesson feedback.

When teachers are asked what makes the workload feel unmanageable, five themes come up repeatedly.

Marking is the most cited driver. The volume is part of the problem; the bigger part is the gap between what teachers are asked to mark and what is genuinely useful to students. Dialogic written marking, where the teacher writes a comment and the student writes a response that the teacher then re-marks, has been a particular flashpoint. The EEF evidence on the time return for this kind of marking is poor.

Planning sits second. The volume is bearable when teachers are working with stable, well-resourced schemes of work, and brutal when each lesson is being designed from scratch. Departments where shared planning is the norm tend to report lower planning hours. The EEF has consistently flagged shared planning as one of the highest-leverage workload moves available.

Administration is third. The administrative burden has grown faster than either marking or planning over the last decade. Data entry, performance management evidence, safeguarding logging, parental communication, behaviour logging, intervention tracking. Each task is justifiable in isolation. The cumulative effect is a substantial chunk of the week.

Meetings come fourth. The general pattern is that schools have more meetings than the system needs, and that the meetings tend to involve information that could have been a clearly written email.

Finally, behaviour management. This one is rarely framed as workload, but the cumulative cognitive and emotional cost of low-level disruption, escalation, and the follow-up paperwork shows up in teacher exhaustion in a way that hours on a timesheet do not capture.

The biggest workload levers are at the school level. A senior team that takes workload seriously can change a teacher's week meaningfully. The DfE workload guidance, particularly the 2018 Workload Review Group reports, identifies four high-leverage areas.

First, marking policy. Schools that have moved from prescriptive written marking to a model that emphasises whole-class feedback, live marking, and self-assessment have consistently reported workload reductions of several hours a week per teacher without measurable impact on student outcomes. The shift requires the policy to be rewritten, the senior team to back the change, and book scrutiny to look for evidence of feedback rather than teacher annotations.

Second, shared planning and resources. Departments that build common schemes of work and shared resource banks reduce per-teacher planning time substantially. The blocker is usually cultural, a perception that good teachers plan their own lessons. The evidence is fairly clear that quality of resources matters more than provenance.

Third, meetings. The blunt move is to halve the number of meetings, replace half of those with clearly written emails, and use the time saved for planning or marking. Schools that have tried this report significant workload gains and no degradation in coordination.

Fourth, data and admin. Reducing the volume of internal data collection is the biggest single workload move available to most schools. Tracking grades three times a term generates hours of work per teacher per cycle, and most of the data is never used in a way that justifies the input cost. Schools that have cut data drops to one or two per year have reported substantial reductions.

Most of the levers above sit with schools. Some sit with individual teachers, and they are worth naming clearly.

One of the highest-impact moves available is rethinking marking. If your school's marking policy permits whole-class feedback, live marking, and self-assessment, using those approaches is the highest-impact change available. The EEF evidence is fairly clear that whole-class feedback delivered well produces equivalent or better outcomes to detailed individual marking, at a fraction of the time cost. If the policy does not permit this, the conversation is with the head of department, not with yourself at midnight.

The second move is being deliberate about what gets marked at all. Not every piece of student work needs detailed feedback. Low-stakes practice and retrieval starters can be self-marked or peer-marked, with the teacher's attention reserved for the work where feedback will actually change what the student does next.

The third move is planning across the year, not the week. Teachers who plan a unit in detail at the start, with all the resources and assessments ready to go, tend to spend much less time per lesson than teachers who plan lesson by lesson.

A fourth move, harder to operationalise but real, is saying no. The teachers with the longest careers in the profession are usually not the ones who took on everything they were asked to do. They are the ones who said no to enough of the extras to protect the core of the job. Sustainable teaching involves declining some things you could in principle do.

Marking deserves its own section because it is one of the biggest workload drivers and the area where the gap between practice and evidence is widest.

The EEF's guidance on written marking, published in 2016 and broadly held up since, makes three points. First, the quality of feedback matters more than the quantity. Detailed written comments do not consistently produce better outcomes than briefer, more focused feedback. Second, feedback needs to result in action by the student; marking that the student reads and ignores is wasted effort. Third, whole-class feedback, where the teacher identifies common errors and addresses them in the lesson, is often as effective as individual written marking and much faster.

The practical implication is that most teachers should be marking less, and the marking they do should be more targeted. A short note identifying a misconception, a model paragraph for the class to compare their work against, a question the teacher addresses in the next lesson. These tend to work as well as a page of written comments, and they take minutes rather than hours.

Early career teachers, particularly in their first two years, tend to experience workload most acutely. They are still building the planning library that experienced teachers lean on, still developing the judgement about what is essential, and often the most reluctant to push back on extras.

The DfE's Early Career Framework, introduced in 2021, includes protected non-contact time and structured mentoring. Schools where the ECF time is genuinely protected tend to retain their early career teachers. Schools where the ECF time is the first thing to be eaten when cover is needed tend not to.

For early career teachers themselves, the most important moves are using available scheme of work resources rather than building from scratch, leaning on the mentor for permission to skip things, and recognising that the second year is easier than the first. The first two years are unusually demanding and the goal is to survive them with enough left in the tank to enjoy year three.

It would be wrong to write about workload without acknowledging that some things have improved. The DfE Workload Reduction Toolkit, the union workload campaigns, and the marking guidance produced by the EEF have moved the needle. Average teacher hours, while still too high, are lower than they were a decade ago. The most egregious practices around triple marking have largely disappeared from policy, if not always from practice.

The honest assessment is that the system is in a slightly better place than it was, and not in a good place yet. Wellbeing indicators are not improving in line with the hours data. The work to be done is the work that was started in 2016 and 2018, continued more consistently across schools. None of this requires new research or new policy.

Pulling the ideas together, here is a short checklist for reviewing workload at the department or individual level. None of it is a magic fix; the value is in the cumulative effect of multiple small changes.