Student Radar
All articles
Assessment articlesAssessment practice

Compare, respond, revisit: a light-touch feedback routine

Use a compare, respond, revisit feedback routine to check pupil action, later independent evidence, adjustments and human review without heavy marking.

Founder, Student Radar

Sources last checked .

  • Teacher feedback
  • Formative assessment
  • Pupil response
  • Teacher workload

Use feedback as a short evidence cycle: compare the pupil's work with the intended learning, give time and access to respond, then revisit the same idea in a later independent task. The final check matters. A comment in a book shows that feedback was given; it doesn't show what the pupil understood, used or retained.

This classroom routine is for teachers, assessment leads, SENCos and school leaders in England. It works with spoken, written, modelled or peer feedback. Choose the form that fits the subject, pupil and task instead of creating a school-wide handwriting endurance event.

Compare one piece of work with one intended learning point

Start with purpose. Name the curriculum point and the teaching decision the evidence could inform. Compare a work sample with the task, success criteria or model that pupils actually received. Don't begin by counting comments, ticks or colours.

Keep enough provenance to interpret the comparison: pupil, subject, date, task and version; curriculum exposure; response conditions; support and reasonable adjustments; feedback mode; missing work; and the teacher making the judgement. A polished sample without those details can look precise while answering the wrong question.

The Department for Education's marking review describes effective marking as meaningful, manageable and motivating. Its workload report also places oral feedback, classroom dialogue and reading pupils' work within the same professional repertoire. The method serves the purpose; volume isn't a proxy for quality.

Three tactile paper evidence stations showing a work sample, a pupil response and a later independent revisit
Editorial illustration: the first work sample, the pupil's response and a later revisit remain separate so each part of the evidence cycle can be checked.

Respond while the feedback can still change something

Give the pupil a small, visible action. They might correct one step, explain a choice, compare two examples, redraft a sentence, rehearse orally or show the idea through their usual communication method. Build this into lesson time and model how to use the feedback.

Preserve ordinary support and adjustments. Record whether the pupil had the expected prompt, equipment, language support, processing time or alternative response mode. A missing response stays missing. It must not become evidence of low ability, poor effort or refusal.

The Education Endowment Foundation's feedback guidance asks teachers to plan how pupils will receive and use feedback. Its evidence review also cautions against treating feedback as one uniform intervention. Effects vary with the content, context and way feedback is used.

Revisit later, with the answer no longer sitting beside the pupil

Return to the same underlying idea in a later task. Keep the curriculum demand comparable, remove any temporary correction prompt and retain the pupil's normal support. The revisit may happen in the next lesson or after a longer interval when retention is part of the question.

Compare the first sample, response and revisit. Record what changed, what remained uncertain and what evidence is absent. Improvement is compatible with the feedback helping, but it does not prove causation. Teaching, practice, task familiarity, attendance, support, maturation and measurement error may all contribute.

Use a six-line record, then stop recording

  1. Purpose: the curriculum point and decision this cycle informs.
  2. First sample: source, date, task version, conditions, support and gaps.
  3. Feedback: the precise point made and how it was communicated.
  4. Pupil response: what the pupil did, said or corrected, with access noted.
  5. Revisit: the later independent evidence and any missingness.
  6. Review: record the class teacher by name, add a subject or assessment lead where moderation is needed, set the review date, state what could change the judgement and explain how the pupil, parent or another professional can question or correct the record.

Sample the record for a real teaching question. Don't require it for every book or every comment. In small classes or narrow groups, report counts and evidence gaps carefully; avoid percentages that identify pupils or make an unstable difference look generalisable.

Walk through one made-up Year 6 science example

These details are invented and combined to show the routine. They describe no pupil, school or Student Radar customer. A teacher samples explanations of why evaporation happens faster in some conditions. One pupil names heat but does not connect it to particle movement.

The teacher compares the response with the lesson model and gives spoken feedback plus the pupil's usual visual vocabulary support. The pupil revises one explanation aloud, then records it in writing. The teacher notes the task, support, feedback mode and an authorised absence during an earlier lesson in the sequence.

Three days later, a different investigation asks for the same causal link without the correction prompt. The pupil explains the link using the usual visual support. The class teacher records that the evidence is stronger, not that the pupil has acquired a fixed level or that feedback alone caused the change. The science lead reviews the small sample on Friday.

Paper-cut work, attendance, support and pupil response paths meeting at an adult-held review lens with one evidence gap left open
Editorial illustration: work, attendance and support context meet at a human review while missing evidence and a route back to the record stay visible.

Connect context only when it changes the professional question

In the example, attendance and adjustment records change how the teacher interprets the first sample. They don't diagnose a need, assign a group, trigger a referral, calculate risk or make a high-stakes recommendation. The teacher and science lead retain the judgement.

Student Radar's Digital Markbook can keep dated outcomes and evidence close to class summaries. Where a formal assessment judgement needs review, Assessment Entry supports submitted, in-review, confirmed and returned states with a reviewer record.

These surfaces can keep the source and review state findable. They cannot determine whether feedback worked, interpret missing evidence or replace the teacher responsible for the next lesson. For the evidence before feedback, use the whole-class hinge routine. When later evidence conflicts, use the conflicting-results review. To inspect the workflow, request an assessment-focused walkthrough.

Sources and further reading

Back to all articles