Why does a school's institutional memory reset every September?

Every September, a school begins again. New timetables, new form groups, new teachers meeting students they have never taught. The routines are familiar and, in many ways, healthy. But something less visible also resets: the accumulated understanding that staff built about individual students over the previous year. That understanding lives in people's heads, in scattered notes, in informal conversations. When a teacher moves on or a student changes section, it largely disappears.

The question worth asking is whether that reset is necessary, or simply the result of not having a better system. Institutional memory, meaning the organised record of what a school has learned about its students over time, is not the same as an MIS or a safeguarding log. It is the richer, more contextual understanding that shapes professional judgement: the pattern a specialist noticed two years ago, the family context a tutor recorded, the student's own account of how they learn. Losing it is not inevitable. It is a structural choice, even if an unintentional one.

This post argues that cumulative understanding is one of the most practical assets a school can build, and outlines what it takes to preserve it across terms, transitions and leadership changes.

• In short
• A school's understanding of its students is built from multiple sources: student voice, teacher observation, formal assessment, specialist insight and family context. When these are not organised, the understanding is fragile and personal rather than institutional.
• Annual resets waste professional effort and can lead to avoidable inconsistency in how students are understood and supported.
• Cumulative evidence, kept at the right level of detail for each audience, supports stronger decisions without replacing professional judgement.
• The people who need evidence are not all the same: a class teacher, a section principal, a Head of School and a board member need different readings of the same underlying information.
• The goal is not surveillance. It is continuity of understanding in service of the student.

What actually resets when September arrives?

The reset is rarely deliberate. It happens because understanding is stored in people rather than in systems. A tutor who knew a student well leaves, and their notes, if they existed at all, do not transfer in any organised way. A student who struggled in Year 8 arrives in Year 9 as a blank page to their new teachers. A specialist who worked with a student for one term writes a report that is filed and rarely revisited.

The problem is compounded by transitions. Moving from junior to senior school, changing from one curriculum track to another, arriving mid-year: each of these is a moment where context is thin and the student must establish themselves again from scratch. For many students that is fine. For some, the loss of accumulated understanding by the adults around them is a meaningful disadvantage.

This is not a criticism of teachers or admissions staff. It is a systems observation. When there is no shared place to keep cumulative evidence, continuity depends entirely on individual memory and informal conversation. That is not a sustainable foundation for institutional judgement.

Why does this matter for how a school makes decisions?

Professional judgement is at its strongest when it is informed, consistent and traceable. When the same case is assessed by different people using different information, decisions can vary in ways that have nothing to do with the case itself. Kahneman, Sibony and Sunstein (2021) describe this as noise: the unwanted variability in judgements that should, in principle, be the same. Structured, consistent processes, including consistent access to the same evidence, can reduce it. (Noise: A Flaw in Human Judgment, Little, Brown Spark.)

In a school context, noise can mean that a student is understood very differently by two equally experienced teachers, not because the student has changed, but because one teacher has access to context the other does not. Decisions about setting, support, opportunity and communication then rest on an uneven base. The antidote is not to remove professional judgement. It is to make sure that judgement is working from shared, organised evidence.

There is a further argument from the student's side. Cook-Sather (2006) argues that young people have unique perspectives on their own learning and schooling, and that their insights warrant the attention and the response of adults. (Curriculum Inquiry, 36(4).) If student voice is worth gathering, it is worth keeping. A system that records what a student says this term but discards it by next September is not really listening.

What are the five sources of evidence a school already holds?

A school does not need to create new evidence from nothing. The understanding already exists, distributed across different people and roles. The five sources are:

• Student voice: what the student says about how they learn, what they find difficult, what matters to them.
• Teacher observation: what teachers notice in class, over time, beyond formal marks.
• Formal assessment: test results, reports and tracked progress data.
• Specialist insight: input from learning support, counsellors, coaches and external professionals.
• Family context: what parents and guardians share that shapes how a student is understood.

These five sources, taken together, are richer than any single one. The challenge is that they are held by different people, recorded in different places and consulted at different moments. Bringing them into a coherent picture, without merging them into a single flattened score or label, is the structural problem worth solving.

Who needs to see what, and at what level?

Not everyone needs the same reading of the evidence. A class teacher needs depth on the individual students they teach. A year group leader needs a sense of the cohort. A section principal needs to lead delivery and answer to the Head. A Head of School needs to brief and assure the board. The board needs to challenge and be assured, not manage operations. A finance committee needs to understand the investment implications of educational priorities, not see individual student records.

This layering matters. When evidence is presented at the wrong level, it either overwhelms or obscures. A board that receives student-level detail is being asked to manage rather than govern. A teacher who receives only aggregate data cannot act on it. The usefulness of cumulative evidence depends on whether it reaches the right person in the right form.

Audience	Primary need	What evidence supports
Class teacher	Individual student understanding	Differentiation, feedback, conversation with families
Year group or section leader	Cohort patterns and transitions	Support deployment, flagging, communication
Head of School or Principal	Delivery briefings and strategic picture	Leading the school, answering to the board
Board	Assurance and challenge	Governance, not operational management
Finance committee	Investment implications of educational priorities	Resource decisions grounded in programme evidence

What happens to the evidence when a student leaves or changes class?

This is where most systems fail. When a student changes class, their record should not reset. When they move from junior to senior school, the understanding built in earlier years should be available, at the appropriate level, to the people who now work with them. When a student leaves the school, their record should be archived, not deleted. They belong to the years they were there.

This is not about surveillance or indefinite data retention for its own sake. It is about the school being able to answer a straightforward question: what did we know about this student, when did we know it, and what did we do with that knowledge? That question is sometimes asked by families, sometimes by inspectors, sometimes by the school's own leaders reviewing a decision. A school with no organised institutional memory cannot answer it reliably.

Questions worth asking before your next leadership meeting

• If a key member of staff left tomorrow, what understanding of their students would leave with them, and what would remain?
• When a student arrives in a new year group, what does their new teacher actually know about them before the first lesson?
• How does student voice gathered in one term inform decisions made two terms later?
• When the board asks about educational priorities, what evidence underpins the Head's answer?
• Is the school's understanding of its students growing more sophisticated over time, or restarting each year?
• Could two senior leaders, given the same student record, reach the same professional conclusion from the same evidence?

FAQ

Is cumulative institutional memory the same as a student information system?

No. A student information system, or MIS, typically holds administrative records: enrolment, attendance, timetables and formal grades. Institutional memory is the richer professional understanding built from observation, conversation, specialist input and student voice over time. The two can sit alongside each other, but they serve different purposes. An MIS tells you what happened. Institutional memory tells you what was understood, by whom, and when.

Does keeping more evidence about students raise safeguarding concerns?

Organised evidence and safeguarding are related but distinct. A safeguarding system of record has its own legal and procedural requirements, and those should not be conflated with the broader institutional memory described here. The practical principle is that evidence should be kept at the appropriate level, shared with those who have a professional need, and governed by the school's data policies. More organisation does not mean more risk; in fact, disorganised records are their own safeguarding vulnerability.

How do you present this kind of evidence to a board without overloading them?

The board's role is assurance and challenge, not operational management. That means the evidence they receive should be a derived reading: patterns, trends and the professional judgements leadership has drawn from the fuller picture, not raw student-level data. A Head briefing a board should be able to say what the school knows, what it has done with that knowledge, and what it is watching. That is a leadership function, not a data dump.

What makes student voice genuinely useful rather than a box-ticking exercise?

Student voice has value when it is taken seriously enough to inform decisions and when students can see that their input has made a difference. Cook-Sather (2006) argues that student perspectives on learning and schooling warrant the attention and response of adults. (Curriculum Inquiry, 36(4).) The structural question is whether student voice is gathered once and filed, or kept as part of an ongoing record that shapes how adults understand and respond to that student over time.

Does cumulative evidence risk labelling students unfairly?

This is a genuine concern and worth taking seriously. The answer lies in how the evidence is used. Evidence that is traceable, dated and attributed to a specific source can be read critically: a teacher can see that an observation was made three years ago by a different member of staff in a different context, and weigh it accordingly. That is preferable to an undated, unattributed impression that exists only in informal conversation and cannot be examined or questioned.

How Evalent supports this

Evalent Intelligence is built around the principle that the understanding already exists in a school, distributed across teachers, specialists, families and students themselves. It organises evidence from those five human lenses into a longitudinal record that does not reset when a student changes class or a teacher moves on, and presents it at the right level for each audience, from the individual student profile through to the board and finance committee. Where a school also uses Evalent Admissions or Course Selector, that evidence can feed into Intelligence automatically.

To see how it works in practice, visit Evalent Intelligence and book a demo from the page.