The Tutor Handbook Vol No.0075 | The Coaching Receipt

The Tutor Handbook · Volume 0075 · Series ID THB-0075

The Tutor Handbook: Complete series index. This volume is the professional-learning companion to Volume 0070 | The Rehearsal-to-Live Gate. Volume 0070 asks whether professional judgement practised in cases and rehearsals can safely cross into real tutoring. Volume 0075 asks what happens after a real lesson has been observed and the tutor receives coaching: how do we know the coaching changed live teaching rather than merely producing a good debrief?

The observation ends.

The coach says, “You answer too quickly after the learner hesitates. Give her longer to think before you rescue the step.”

The tutor agrees immediately.

“Yes. I can see that.”

They discuss two moments from the lesson. The tutor explains why waiting matters. They rehearse a better response. The debrief is intelligent, respectful and specific.

Three days later, a learner pauses over a new algebra problem.

The tutor waits one second, feels the silence, and says, “Think about what we did last time. Which method is this?”

The coached habit did not survive the transition back into live teaching.

The Coaching Receipt is observable evidence that professional feedback changed a tutor’s later teaching move in a real lesson, under conditions where the old habit could plausibly have returned.

This definition sounds demanding because coaching is easy to confuse with conversation. A tutor can be reflective without changing practice. They can agree with a coach without remembering the new move at the decision point. They can execute the move perfectly in rehearsal but lose it when a learner responds unexpectedly. They can change the move while producing no better opportunity for learner thinking. They can even perform the coached behaviour once because the observer is present and then return to the old routine when the observation cycle ends.

Professional learning therefore needs a receipt just as learner feedback does. The receipt is not the tutor saying, “That makes sense.” It is the new teaching decision appearing later, in context, with enough stability that the coach and tutor can learn from what happened next.

The direct answer

A tutor should judge coaching by the next relevant teaching episode. Begin with a specific observed problem of practice, define one small teaching move that should change, make the success condition visible, rehearse when useful, and then return to live tutoring. During the next lesson, observe whether the tutor notices the relevant moment and enacts the new move without needing the coach to trigger it. Then inspect the learner opportunity created by that move: did the learner actually receive more thinking time, a clearer prompt, a better feedback opportunity, a fairer chance to explain, or whatever the coaching target was meant to create? Repeat across enough sessions or conditions to distinguish a one-off performance from a changed professional routine.

The National Student Support Accelerator at Stanford recommends ongoing observation, feedback and individualized coaching for tutors, with coaching focused on small areas aligned to tutors’ goals and demonstrated needs. Its early-literacy professional-learning framework also emphasises practice-based learning, models, rehearsal, enactment with students and later analysis. The framework is particularly relevant because it is tutor-specific, but its direct scope is early literacy tutoring and provider design, so it should not be treated as proof that one coaching protocol works identically across every subject, age or private-tuition context.

The wider evidence base on teacher coaching is encouraging but also needs careful transfer. Kraft, Blazar and Hogan’s 2018 meta-analysis of 60 causal studies reported sizeable average effects on instructional practice and smaller positive average effects on student achievement, with much of the evidence coming from literacy coaching for pre-kindergarten and elementary teachers in the United States. The review also found weaker effects when programmes scaled. That is useful evidence that coaching can influence practice; it is not a licence to assume that any coaching conversation automatically improves a tutor or a learner.

1. What this volume owns

This volume owns the receipt after tutor coaching. It begins after a real lesson, real observation or sufficiently representative live-teaching sample has generated professional feedback. The question is whether that feedback becomes changed tutoring practice.

It does not own pre-service rehearsal. Volume 0070 already owns the transition from cases and rehearsals into first live enactment. It does not own the learner’s feedback uptake; Volume 0072, The Feedback Receipt, owns that. It does not create a universal teacher-evaluation rubric or an accreditation system for tutors.

The job here is narrower and more useful: when a tutor is coached on a specific teaching move, what evidence shows that professional learning crossed from the debrief into the next real lesson?

2. Agreement is not professional change

Tutors often agree with good feedback because it is genuinely persuasive. Video or observation can make a pattern visible that was hard to notice from inside the lesson. A coach may point out that the tutor answers their own questions, gives three prompts where one would do, praises speed while claiming to value reasoning, or spends most of a three-student lesson speaking to the strongest student.

Recognition matters. A tutor cannot deliberately change a pattern they cannot see. But recognition is only the beginning. Professional habits operate under time pressure. The tutor is reading the learner, managing materials, tracking two other students, remembering the lesson goal and deciding whether to intervene. The old move can return even after the tutor sincerely agrees that it should change.

The receipt must therefore come from behaviour in the conditions where the habit normally operates.

3. Reflection is not yet professional change

A reflective tutor can explain the lesson beautifully after it ends. They can identify the precise moment where they over-prompted, describe what they should have done, and connect the mistake to learner independence.

That reflection is valuable professional knowledge. It becomes a coaching receipt only when the tutor can act differently the next time a similar moment arrives.

This distinction protects coaching from becoming an intellectual performance. Tutors who are articulate, experienced or comfortable with educational language can sound highly developed in a debrief while maintaining the same live habits. Conversely, a quieter tutor may struggle to verbalise a theory of change while implementing the coached move reliably.

Professional learning should value explanation, but live enactment remains the stronger receipt.

4. Rehearsal is not yet the coaching receipt

Rehearsal is one of the most useful bridges between feedback and live practice. Stanford’s practice-based tutor framework recommends models, rehearsal, feedback, enactment and later analysis. Rehearsal lets the tutor slow down a teaching moment, try new wording, practise waiting, test a sequence and receive immediate coaching without using a learner as the place where every professional experiment first occurs.

But rehearsals simplify the world. The “student” in the rehearsal may respond cooperatively. The tutor knows which skill is being practised. There are no real stakes, competing learner needs or surprise misconceptions unless the facilitator deliberately introduces them.

Rehearsal shows that the tutor can enact the move under practice conditions. The Coaching Receipt asks whether the move survives live complexity.

5. The coaching target should be small enough to observe

“Improve questioning.” “Be more student-centred.” “Differentiate better.” “Build confidence.” These may describe worthwhile ambitions, but they are poor coaching targets because the next observation cannot easily determine what changed.

A better target names a tutor move at a decision point.

After asking a substantive question, wait long enough for the learner to begin a genuine response before supplying a method cue.
When a learner gives a wrong answer, ask one diagnostic question before explaining the solution.
During a three-student task, collect an independent first attempt from all three before discussing one learner’s answer publicly.
When giving feedback, identify one actionable criterion and require a fresh attempt before adding another correction.
At the end of the lesson, ask the learner to state the next independent action rather than summarising it for them.

Small does not mean trivial. A small move can control a large amount of learner thinking. It is also observable enough to generate a receipt.

6. The target should come from evidence, not coach taste

Coaches have preferences. One values discussion. Another likes explicit routines. One prefers quiet wait time. Another prefers fast, tightly scaffolded practice. Professional coaching becomes dangerous when stylistic preference is presented as universal pedagogy.

The target should be tied to an observed learning problem or an explicit instructional routine. “You should ask more open questions” is weak if the session target requires rapid fluency practice. “Your current prompts tell the learner which operation to perform before she has a chance to select it” is stronger because it identifies a mismatch between the tutor move and the learner job.

The coach should be able to answer: what learner opportunity is being lost, what tutor move contributes to that loss, and what changed move should restore the opportunity?

7. Begin with descriptive observation before evaluation

Professional feedback becomes more precise when it begins with what happened rather than a label.

“You are too controlling” is evaluative and broad.

“In the six minutes I observed, you asked Alicia four method-selection questions. On three of them, you supplied the method after less than two seconds of silence. Alicia then executed the method accurately.”

The second description creates a professional problem that can be investigated. Was the wait genuinely too short? Did Alicia need the cue? Was method selection the intended target? Would longer waiting produce productive thinking or simply leave her stalled?

Stanford’s practice-based framework similarly recommends descriptive analysis of what tutors and students can do rather than broad labels such as “high” or “low.” Description creates a cleaner bridge from observation to coaching target.

8. Coaching needs a learner-facing reason

A coaching move should not exist only to make the tutor look more polished. The reason should eventually connect to the learner.

If the coach asks for longer wait time, the learner-facing reason may be to preserve genuine method selection before prompting. If the coach asks the tutor to shorten explanations, the purpose may be to create more learner retrieval and production. If the coach asks the tutor to collect three independent attempts before discussion, the purpose may be to protect diagnostic information and prevent one student from carrying the other two.

The learner-facing reason also prevents fidelity from becoming mechanical. The tutor is not merely copying the coach’s behaviour. They understand the function well enough to adapt the move when the learner or subject changes.

9. The coaching chain has six stages

Observe: capture a representative teaching moment.
Interpret: connect the tutor move to a learner opportunity or instructional goal.
Specify: define one changed tutor move that is small enough to enact and observe.
Prepare: model, discuss or rehearse the move when useful.
Enact: use the changed move in a live tutoring session.
Read the receipt: inspect whether the move occurred, what learner opportunity it created, and whether it should be kept, adapted or rejected.

Skipping the enactment stage turns coaching into advice. Skipping the receipt turns changed behaviour into a ritual whose learner value is assumed rather than observed.

10. The first receipt is tutor behaviour

The first coaching receipt is simple: did the tutor do the coached thing when the relevant moment appeared?

If the target was waiting before cueing, did the tutor wait? If the target was asking one diagnostic question before explaining, did the tutor ask it? If the target was allowing an independent first attempt, did the tutor preserve that attempt?

This matters because coaching can fail before learner outcomes are even considered. If the tutor never enacted the change, a later lack of learner improvement cannot fairly be used to judge the coaching idea. Implementation is part of the causal chain.

11. The second receipt is the learner opportunity

A changed tutor behaviour is not enough. The move should create the learner opportunity it was designed to create.

The tutor waits longer, but fills the silence with facial signals that announce whether the learner is correct. The formal behaviour changed; the learner still does not own the decision.

The tutor asks one diagnostic question, but the question is so leading that it reveals the method. The coach can tick the box “asked diagnostic question,” but the learner opportunity remains weak.

The receipt therefore needs functional reading: did the changed tutor move actually alter what the learner had to think, choose, explain or carry?

12. The third receipt is learner response

Once the learner receives a better opportunity, what happens?

The learner may produce a stronger explanation, reveal a misconception that was previously hidden, attempt a method independently, become productively stuck, ask a more precise question or show that the coach’s hypothesis was wrong.

All of these can be useful receipts. Coaching should not be judged only by whether the learner immediately succeeds. Sometimes a better tutor move makes difficulty more visible because the old prompt had been hiding it.

That can be an improvement in teaching even before it becomes an improvement in learner performance.

13. A better teaching move can initially make results look worse

Suppose the tutor stops supplying method cues. The learner’s accuracy falls on the next mixed set. Superficially, the coaching made teaching worse.

But the new condition has revealed that method selection was weaker than the tutor thought. The old teaching routine generated correct answers while concealing dependence. The coached move has produced better diagnostic evidence.

The next decision is not to celebrate lower marks. It is to use the newly visible weak link. Coaching quality should be judged by whether it improves the learning system, not whether every immediate metric rises.

14. A better teaching move can improve the lesson without proving coaching caused the learner gain

A learner improves after the tutor adopts a new questioning routine. School also reteaches the topic. The learner studies more independently that week. A parent changes the homework environment.

Do not force all credit onto coaching.

Volume 0065, The Concurrency Problem, still applies. The coach can say with stronger confidence that the tutor changed the teaching move if the behaviour was observed. The coach can say the learner received a different opportunity. Causal claims about achievement should remain proportionate to the wider evidence.

15. One observed change is not yet a stable coaching receipt

The observer returns. The tutor remembers the target and performs it beautifully. Was practice changed?

Perhaps. The presence of the coach can itself cue the behaviour. A stronger receipt appears when the tutor uses the move in later lessons, with different students, or in a moment the coach did not preselect.

This is professional transfer. The tutor has moved from “I can perform the coached routine while being observed for it” to “I notice when this move is needed and use it as part of normal teaching.”

The required evidence should match the importance of the change. A minor wording adjustment may need little proof. A major change in tutoring structure deserves a broader stability window.

16. The tutor must learn the trigger, not only the move

Professional expertise involves knowing when to use a move.

A tutor can learn to wait after every question and become painfully slow even when the task is simple recall. They can learn to ask diagnostic questions and interrogate a learner who simply needs a clear explanation. They can learn to withhold help and turn productive struggle into pointless delay.

The coaching target should therefore contain a trigger condition: “When the learner has enough prior knowledge to attempt method selection, do not supply the method before observing the first choice.” “When the learner’s error could come from more than one cause, ask one discriminating question before explaining.”

The Coaching Receipt becomes stronger when the tutor identifies the right moment without the coach announcing it.

17. Bite-sized targets protect interpretability

Stanford’s coaching framework highlights “bite-sized” learning aligned to tutors’ goals and needs. There is a practical reason for this beyond convenience. If the coach asks the tutor to change questioning, pacing, feedback, grouping and homework design at once, neither person can easily tell which new move helped or which one failed.

This mirrors Volume 0067, The Change Bandwidth. Tutors themselves have change bandwidth. A professional-learning plan can overload an educator just as a learning plan can overload a student.

One important move, repeatedly enacted and reviewed, often produces better professional information than five worthy intentions.

18. Directive coaching has a place

A novice tutor may simply not know the routine. They may need a coach to model how to elicit an independent first attempt, how to run a short retrieval check or how to correct a reading error without taking over the whole word.

In such cases, directive coaching can be efficient: explain the move, show the move, rehearse it, then observe live enactment. Stanford’s tutor framework describes directive approaches as one part of a wider coaching spectrum and notes that coaches may use different approaches at different times.

The receipt remains the same. Did the tutor later enact the move appropriately, and what learner opportunity did it create?

19. Collaborative coaching has a place

An experienced tutor may understand the pedagogy and still face a complex case. Coach and tutor can inspect the evidence together, compare interpretations and design a small change.

Here the coach is not merely transmitting a routine. They are helping the tutor think more precisely about the learner and the teaching problem. The coaching receipt may therefore include both changed teaching and improved professional judgement: the tutor begins identifying the same class of problem independently in later cases.

Volume 0069, The Evidence Conference, provides the stronger protocol when two professionals genuinely disagree about what the evidence means.

20. Facilitative coaching has a place

Sometimes the tutor already has the knowledge needed to improve but has not connected it to the observed pattern. A coach can ask questions that help the tutor surface their own interpretation and select the next move.

“What were you hoping Alicia would do during that pause?” “What did your prompt make easier?” “Which part of the decision remained hers?” “What would you want to see if you waited longer?”

This can strengthen tutor agency, but facilitation should not become an ideology. If the tutor lacks a necessary skill or misconception is clear, refusing to provide expert guidance can waste time. Coaching method should respond to professional need just as teaching responds to learner need.

21. Trust matters because observation creates vulnerability

Live tutoring is personal work. A tutor is making rapid decisions in front of a learner. Observation can feel evaluative even when the stated purpose is development. Stanford’s framework places strong emphasis on trusting coach–tutor relationships and on feedback grounded in care, respect and shared commitment to learner growth.

Trust does not mean avoiding difficult feedback. It means the tutor can expose uncertainty without assuming that every mistake will become a judgement of professional worth.

A coaching culture that punishes visibility creates predictable data problems: tutors perform safe routines during observations, hide difficult cases and avoid admitting when a strategy failed. The receipt then measures observation performance rather than professional learning.

22. Psychological safety is not the same as low standards

A coach can value the tutor as a person while holding a high standard for teaching. “This learner did not receive a real chance to answer before you supplied the method” is a clear professional observation. It does not need to become “You are a bad tutor.”

The distinction is important because identity-level criticism is difficult to act on. Behaviour-level feedback can generate a changed move and therefore a receipt.

High standards become more usable when the path from observation to action is specific.

23. Coaching should sometimes preserve the tutor’s current move

Observation does not require change. The coach may discover that the tutor’s current routine is functioning well. The learner receives genuine thinking time, prompts are proportionate and the evidence supports the route.

“No change needed here” can be a professional conclusion. Otherwise coaching becomes a novelty machine in which the coach must always produce a recommendation to justify the observation.

The purpose of coaching is improved practice, not visible coach activity.

24. Coaching feedback can be wrong

A coach sees one lesson and concludes the tutor explains too much. The next lesson reveals that the observed student was encountering genuinely new material and needed explicit instruction. The problem was not explanation volume; it was the coach’s inference from a narrow sample.

A tutor should not reject feedback defensively, but neither should professional coaching create blind obedience. Ask what evidence supports the target. Check whether the observed moment is representative. Consider alternative explanations. Trial the changed move when uncertainty remains.

A failed coaching receipt can update the coach as well as the tutor.

25. Coach certainty should match observation quality

A five-minute observation of one learner should support narrower claims than several representative sessions. A coach who sees only the tutor’s strongest group may miss problems that emerge with struggling learners. A coach who observes only examination rehearsal may misread routines used during initial teaching.

The same evidence discipline applied to learners should apply to tutors. One observation is evidence about one performance under conditions. It is not a complete numerical description of professional capability.

Before escalating a coaching target into a broad judgement, gather enough representative evidence.

26. The coach should name the observation condition

Was the session one-to-one or three-student? New topic or revision? Repair, Alignment or Frontier mode? Was the learner familiar to the tutor? Was an examination near? Was the session unusually disrupted?

These conditions do not excuse poor teaching. They help determine how far the observation should travel.

A professional coaching record should preserve enough context that later reviewers do not convert a local observation into a permanent tutor label.

27. Mathematics case: the rescue prompt

Illustrative composite case. A Mathematics tutor is excellent at algebra and fast at spotting errors. During observation, the coach notices a repeated pattern: whenever Alicia pauses, the tutor asks a question that effectively names the method. Alicia then solves accurately.

The coaching target is not “talk less.” It is: on mixed method-selection tasks, preserve the learner’s first method choice before giving a method-level prompt. The tutor rehearses three neutral responses: wait; ask Alicia to state what the question is asking; or ask which features she notices without naming the method.

In the next live lesson, Alicia pauses. The tutor waits. Alicia chooses the wrong method. This looks worse than the old lesson, but the coaching has produced a cleaner learner signal. The tutor can now diagnose why the selection failed.

The first receipt is that the tutor did not rescue the selection. The second is that Alicia had a genuine opportunity to choose. The third is the new diagnostic evidence. Later, after targeted practice, Alicia begins choosing correctly without method prompts. The coaching did not magically cause the whole improvement, but it changed the teaching condition enough for the real weak link to become teachable.

28. English case: the feedback avalanche

Illustrative composite case. An English tutor gives Beatrice detailed feedback on every paragraph: vocabulary, sentence control, evidence, organisation, tone and punctuation. The comments are accurate. Beatrice’s rewrites improve when the tutor sits beside her, but the same weaknesses recur in fresh work.

The coach observes that feedback quantity may be exceeding uptake capacity. The target becomes: choose one high-leverage criterion for the next attempt, make the criterion explicit, then require a fresh application before adding another correction.

During the next lesson, the tutor feels the old impulse to mark everything but limits the live feedback to evidence relevance. Beatrice writes a new paragraph and improves the evidence-explanation link. Grammar errors remain visible but are not allowed to become competing intervention targets in that moment.

The coaching receipt is not a cleaner page of markings. It is changed tutor prioritisation producing a clearer learner next move.

29. Science case: model answers arriving too early

Illustrative composite case. A Science tutor uses strong model answers because they make precise scientific language visible. The coach notices that Ciara often sees the model before she has reconstructed the mechanism herself.

The target is not “stop using model answers.” It is: for mechanism questions where Ciara has enough prior knowledge, collect an independent causal explanation first, then use the model as a comparison tool.

In the next session, the tutor delays the model. Ciara produces an incomplete explanation. The tutor now has evidence of the exact causal link that is missing. After a short repair, Ciara revises and only then compares her wording with the model.

The new move makes the learner’s model visible before polished language can hide it. The coaching receipt is changed sequencing with better diagnostic access.

30. Three-student case: the strongest learner carries the room

Illustrative composite case. In a three-student tutorial, one confident learner answers quickly. The tutor enjoys the energy and often turns that answer into group discussion. The other two students appear engaged, but their independent first thinking is rarely visible.

The coach proposes one structural move: before public discussion of a substantive question, collect a private first response from all three learners. The response may be a written step, selected evidence, diagram or method choice.

In the next lesson, the tutor uses the new routine. One learner’s wrong first step reveals a misconception that would previously have been masked by the strongest student’s answer. The coach and tutor now have a clear receipt: the teaching move changed, learner opportunity changed, and diagnostic information improved.

Again, the point is not that every question needs a private written response. The tutor has learned when the routine is useful: when independent thinking is part of the learning or diagnostic job.

31. Coaching should not turn tutors into scripts

A routine can protect quality, especially for novice tutors. But if coaching produces rigid compliance without understanding, the tutor may use the correct move in the wrong situation.

The long-term goal is adaptive expertise: the tutor knows the mechanism well enough to preserve its essential function while responding to different learners and conditions. The OECD Teaching Compass describes teachers as adaptive experts and co-designers of learning rather than passive recipients of fixed instructions. Although it is a policy framework for teaching more broadly, the principle fits tutor development: professional agency should grow with competence.

A mature Coaching Receipt therefore contains both fidelity and judgement. The tutor can enact the essential move and explain when they would adapt or withhold it.

32. Distinguish core function from flexible form

Suppose the coaching target is to protect an independent first attempt. The core function is clear: the learner must generate evidence before adult explanation contaminates it.

The form can vary. One learner writes. Another sketches. Another speaks after a short silent planning period. In a fast fluency routine, the independent first attempt may happen orally within seconds. In a complex composition task, it may require several minutes.

Coaching should preserve the function while allowing form to adapt. Otherwise the tutor learns a ritual rather than a principle.

33. Video can strengthen the receipt, but privacy comes first

Video can help tutors notice timing, language and interaction patterns that memory misses. Stanford’s practice-based framework discusses video models and tutor video for analysis, while explicitly noting permission requirements and alternatives such as filming only the tutor when appropriate.

A tutoring programme should never treat recording as casually authorised. Student privacy, parental consent, institutional rules and data handling matter. If appropriate permission is not available, use live observation, tutor-only capture where genuinely compliant, representative work samples, transcripts prepared without identifiable information, or structured self-observation.

Professional learning does not outrank learner privacy.

34. AI can support tutor reflection without becoming the coach of record

AI tools may help summarise notes, generate rehearsal cases, compare a transcript against a known routine or prompt a tutor to reflect on questioning patterns. These can be useful professional-learning supports.

But AI-generated observations should not be treated as automatically valid. Transcripts can misattribute speakers. A model can infer intentions that were not observable. It may apply generic pedagogical rules without understanding the learner’s actual target or support needs.

Use AI as an analysis aid where privacy and policy permit, not as unquestioned authority. The coaching receipt still comes from live teaching evidence and professional judgement.

35. The coach should not create dependency either

A tutor can become dependent on coaching in the same way a learner can become dependent on prompts. Every difficult teaching decision is deferred to the coach. The tutor waits for observation before adjusting obvious problems. Professional agency shrinks.

Good coaching should progressively increase the tutor’s ability to observe their own practice, define a problem, select a proportionate change, test it and read the learner response. The coach remains available for harder cases, blind spots and challenge, but does not become the permanent owner of routine judgement.

The final professional receipt is therefore partly self-coaching: the tutor can run a trustworthy improvement loop without needing an external observer for every adjustment.

36. The Coaching Receipt Card

Observed episode: what happened in the live lesson?
Condition: what learner, subject, mode and task were involved?
Learner opportunity lost or weakened: what did the tutor move prevent or obscure?
Coaching target: what one tutor decision should change?
Trigger: when should the changed move be used?
Core function: what learner opportunity should the move protect or create?
Preparation: model, rehearsal, discussion or other support needed?
First live receipt: did the tutor enact the changed move?
Learner-opportunity receipt: did the learner actually receive the intended chance to think or act?
Learner-response evidence: what became visible after the move?
Transfer: did the tutor use the move appropriately in another relevant live context?
Decision: adopt, refine, narrow, reject or replace the coaching target?

37. The coach should wait for the right live moment

A coaching target may not appear in every lesson. If the target concerns handling learner misconceptions, a session without a meaningful misconception is not a failed coaching observation. If the target concerns method selection, a lesson composed entirely of direct instruction may not provide a fair opportunity to observe the move.

This is the professional version of Volume 0068’s Opportunity Check. Do not conclude that a tutor failed to enact a move when the relevant trigger never occurred.

Good coaching waits for a genuine opportunity or deliberately creates a representative practice condition before making a judgement.

38. Coaching receipts need stable enough conditions

A new class format, new curriculum, new tutor role and new coaching target all begin in the same week. If practice looks different, interpretation becomes difficult.

Professional learning has its own concurrency problem. Change one important teaching variable where possible. Give the tutor enough stability to develop the new habit. If the environment is already changing rapidly, narrow the target further or extend the observation window.

The tutor should not be judged against a coaching target under a completely different system from the one in which it was defined without acknowledging the changed conditions.

39. The coach should look for drift after adoption

A teaching move can be adopted and later drift. The tutor begins by waiting for an independent method choice. Weeks later, the wait remains formally present but the tutor starts supplying increasingly leading hints during it.

Do not restart coaching from zero. Use the logic of Volume 0054, The Drift Check. Compare current practice with the function that was originally adopted. Decide whether the variation is a useful adaptation or a return of the old problem.

Professional growth is maintained through periodic evidence, not through permanent surveillance.

40. Coaching receipts should inform future tutor development

When one target stabilises, the next target should not be chosen merely because a rubric has another box. Look at the tutor’s current professional weak link.

A tutor who now protects independent attempts may be ready to improve diagnostic questioning. A tutor who asks excellent diagnostic questions may need to improve how feedback is converted into fresh practice. A tutor who handles one-to-one instruction well may need to improve orchestration across three learners.

Professional learning should therefore have its own route design: one useful next teaching problem at a time, with receipts before escalation.

41. Coaching a Class 0 Homework Helper

For a tutor operating mainly as a Homework Helper, coaching may focus on not taking over task management. A useful target could be: when homework instructions are clear, ask the learner to state the first action and required materials before the tutor organises anything.

The receipt is the tutor transferring one real management decision back to the learner while school completion remains protected.

42. Coaching a Class 1 Explainer

An Explainer may be coached to stop explaining after the learner already has enough understanding to attempt reconstruction. The target is not “explain less” in the abstract. It is to recognise the transition point from explanation to learner production.

The receipt appears when the tutor leaves the next step to the learner and observes whether understanding survives.

43. Coaching a Class 2 Drill Builder

A Drill Builder may be excellent at generating practice but weak at reading when practice has become repetitive. Coaching may focus on introducing variation only after basic execution stabilises, or on reducing volume when the same questions no longer generate information.

The receipt is not a more creative worksheet. It is better alignment between practice condition and the learner’s current reliability problem.

44. Coaching a Class 3 Diagnostic Tutor

A Diagnostic Tutor can become overconfident in elegant explanations of learner difficulty. Coaching may focus on keeping alternative hypotheses alive and asking one discriminating question before committing to a cause.

The receipt is the tutor changing the evidence condition rather than immediately prescribing the preferred repair.

45. Coaching a Class 4 Route Designer

A Route Designer can over-engineer. Coaching may focus on reducing simultaneous changes, clarifying the active priority or protecting a Stability Window long enough to learn from the route.

The receipt is a simpler, more interpretable learning sequence—not a more elaborate planning document.

46. Coaching a Class 5 Performance Coach

A Performance Coach may add pressure too quickly because examination conditions feel urgent. Coaching may target the sequence of performance constraints: preserve accuracy before tightening time, or stabilise section-level recovery before adding full-paper load.

The receipt is calibrated pressure that reveals the next performance bottleneck without destroying the capability already earned.

47. Coaching a Class 6 Learning Architect

The Learning Architect sees many interacting variables. The coaching risk is sophistication without restraint. A coach may need to ask, “Which one learner condition actually needs changing now?”

The receipt is selective intervention: the tutor changes only the layer current evidence justifies while preserving the rest of the learner’s working system.

48. Coaching in Repair mode

In Repair mode, coaching should often sharpen diagnosis, scaffold choice and reintegration. The tutor should not become so focused on the weak link that the lesson turns into endless correction.

A useful coaching receipt may be the tutor collecting a fresh independent attempt after repair instead of assuming the corrected example proves learning.

49. Coaching in Alignment mode

Alignment mode creates strong external pressures from school. Coaching may focus on keeping tuition responsive without becoming a second conflicting school system. The tutor may need to translate school evidence into a focused learning job rather than reacting to every worksheet or upcoming test.

The receipt is coherent support that improves participation in the current educational route while preserving learner independence.

50. Coaching in Frontier mode

Frontier mode can tempt tutors to display expertise through difficult content. Coaching may focus on asking whether extension builds transferable reasoning or merely increases syllabus distance.

The receipt is challenge chosen for intellectual leverage, not prestige.

51. Parent communication about tutor coaching

Parents do not need a transcript of every coaching conversation. They do benefit from knowing that tutoring quality is treated as improvable rather than fixed.

A useful explanation might be: “We review teaching decisions as well as student results. When a tutor changes a practice, we look for the change in later lessons and for the learner opportunity it is meant to create. We do not treat a good debrief as proof that practice changed.”

Do not use professional learning as a marketing performance. The value lies in actual improvement and honest evidence, not in displaying internal observation machinery to families.

52. Learners can contribute to the coaching receipt

Learner voice can reveal teaching changes that observation misses. A student may say, “You give me longer to think now,” “I have to choose the method before you help,” or “You ask me why instead of correcting me immediately.”

That does not prove the change is educationally beneficial by itself. Learners may prefer more rescue or less challenge. But their experience can confirm whether the teaching environment actually changed.

Use learner voice as one source of evidence, not a popularity vote.

53. Coaching metrics can become the wrong target

If tutors know they are being scored on “wait time,” they may exaggerate silence mechanically. If they are scored on “student talk,” they may force discussion when concise direct instruction would be better. If every lesson must contain a prescribed number of checks, checks can become theatre.

Measures are useful when they preserve the professional job. They become dangerous when tutors optimise the indicator rather than the learner opportunity underneath it.

The coach should therefore ask whether the measure still represents the function it was designed to observe.

54. Coaching should distinguish fidelity from rigidity

Some instructional routines have core elements that should be implemented reliably. Others depend heavily on responsive judgement. The tutor needs to know what is core and what is flexible.

For example, preserving an independent first attempt may be core because without it the tutor loses diagnostic evidence. Whether the attempt is written, oral or diagrammatic may be flexible. Asking one discriminating question before explanation may be core in a diagnostic routine, while the exact wording should respond to the learner.

Coaching that cannot distinguish core function from surface form creates brittle tutors.

55. Coaching should include an expiry condition

Once the tutor reliably enacts the move and can adapt it appropriately, intensive coaching on that target should end. Keeping the same target alive forever can freeze the tutor’s professional identity around an old weakness.

The target can move into light monitoring, just as a learner weakness can move into maintenance. Reopen it only if drift or new conditions justify attention.

Professional development should produce release, not permanent remediation.

56. The Coaching Receipt and the Feedback Receipt mirror one another

Volume 0072 asks whether learner feedback changes the learner’s next attempt. This volume asks whether professional feedback changes the tutor’s next teaching attempt.

The structure is strikingly similar:

preserve the first performance;
identify one useful change;
make the criterion clear;
allow rehearsal or guided practice;
require a fresh relevant attempt;
reduce external cueing;
change the context when appropriate;
return later;
move detection increasingly into the performer’s control.

The parallel is not accidental. Professional learning is still learning. Adults also need opportunities to interpret feedback, practise, enact, receive evidence and update.

57. Research foundation: tutor-specific guidance

The Stanford National Student Support Accelerator’s Early Literacy Tutor Professional Learning Framework recommends practice-based formal learning, feedback and individualized coaching, and communities of support. Its coaching section recommends observation and feedback tailored to tutor needs, describes a range of coaching approaches, and encourages systems for tracking tutors’ goals, feedback and progress. Its practice-based section describes models, rehearsal, enactment with students and later analysis.

This is directly relevant to tutor professional learning, but its empirical and implementation context matters. The framework grew from early-literacy tutoring providers and broader teacher-education evidence. A Secondary Mathematics tutor in Singapore should not assume that every early-literacy routine transfers unchanged. The useful transfer is the professional-learning architecture: observe, focus, practise, enact, analyse and adapt.

58. Research foundation: teacher coaching evidence

Kraft, Blazar and Hogan’s meta-analysis, published in the Review of Educational Research in 2018, synthesised 60 causal studies of teacher coaching and found positive average effects on instructional practice and smaller positive effects on student achievement. The authors also highlighted scaling challenges and noted that much of the evidence came from literacy coaching for younger students in the United States.

For this handbook, the most defensible inference is not “coaching always works.” It is that structured coaching is a credible professional-development approach with causal evidence in school teaching, while context, implementation, scale and transfer to private tutoring remain important boundaries.

59. Research foundation: adaptive professional agency

The OECD Teaching Compass, published in May 2025, frames teachers as agents of curriculum change, co-designers of learning and adaptive experts. It also emphasises professional competencies, agency and collaborative professional learning.

This provides a useful boundary for coaching. The end state is not a tutor who follows more instructions. It is a tutor whose professional judgement has become more accurate, more responsive and more capable of acting independently while remaining answerable to evidence.

60. Evidence limits

The exact Coaching Receipt protocol in this article has not been validated as a named intervention. It is a tutor-facing synthesis built from tutor professional-learning guidance, teacher-coaching evidence and the wider logic of practice-based learning and formative improvement.

Do not infer a fixed coaching frequency from this article. Stanford reports wide variation among tutoring providers, and the appropriate observation cycle depends on tutor experience, programme structure, target complexity, resources and risk. Do not infer that video is required. Do not infer that every coaching target should be directive. Do not infer learner achievement from one changed tutor behaviour.

The evidence supports disciplined professional learning, not one universal coaching script.

61. Tutor audit

What exact teaching episode generated this feedback?
What learner opportunity was weakened or lost?
What one tutor move should change?
When should that move be triggered?
What is the core function of the move?
Do I understand why it matters, or only what I was told to do?
Do I need a model or rehearsal before using it live?
What would count as my first live receipt?
How will I know whether the learner received the intended opportunity?
What learner response would confirm, complicate or reject the coaching hypothesis?
Can I use the move in another relevant context without the coach prompting me?
When will this coaching target be stable enough to retire?

62. Coach audit

Am I describing evidence before applying a label?
Is the observation representative enough for the claim I am making?
Have I connected the tutor move to a learner-facing reason?
Is the target small enough to enact and observe?
Am I coaching a real need or my personal style preference?
Does this tutor need directive, collaborative or facilitative support right now?
Have I clarified the trigger as well as the move?
What should remain flexible?
What privacy boundaries govern observation or video?
Did the tutor actually enact the change?
Did the change create the intended learner opportunity?
What did the learner response teach us about the coaching target?
Am I willing to revise my coaching hypothesis if live evidence disagrees?
When should I reduce coaching support and return more professional judgement to the tutor?

63. Programme audit

Do tutors receive ongoing professional support rather than one-off induction only?
Are observations developmental as well as evaluative?
Can coaches focus on small, evidence-linked targets?
Is there a way to preserve coaching goals and receipts over time?
Do tutors have safe opportunities to rehearse difficult moves?
Are learner privacy and consent protected in any recording or work-sample process?
Does coaching preserve tutor agency as expertise grows?
Are coaching metrics used as evidence rather than targets to game?
Can a coaching target be retired when it stabilises?
Does the programme learn when several tutors struggle with the same routine, rather than treating every issue as an individual defect?

64. What the Coaching Receipt is not

It is not the tutor saying, “I agree.”
It is not a thoughtful reflection note.
It is not a perfect rehearsal.
It is not one coached performance while the observer is present.
It is not a checklist score detached from the learner opportunity.
It is not blind fidelity to a coach’s preferred style.
It is not proof that coaching alone caused later learner achievement.
It is not surveillance disguised as professional learning.
It is not permanent dependence on the coach.
It is evidence that professional feedback changed a relevant teaching decision in live practice and that the changed decision created the learner opportunity it was intended to create.

65. The ethical standard

Tutors have the same right to intellectually serious feedback that learners do. They should not be reduced to labels, polished by style preferences or kept dependent on a more senior adult. At the same time, “professional autonomy” cannot mean that every tutoring habit is beyond observation or evidence.

The coach owes the tutor a fair observation, a proportionate claim and a useful path to action. The tutor owes the learner a willingness to change when the evidence shows that a teaching habit is limiting learning.

Coaching has done its job when professional feedback becomes a better live teaching decision, that decision creates a better learner opportunity, and the tutor can eventually recognise and run the improved move without needing the coach to stand beside it.

Evidence and connected reading

Final compression

Observe a real teaching problem. Describe before labelling. Connect the tutor move to a learner opportunity. Choose one small professional target. Name the trigger. Rehearse when useful. Enact it live. Check whether the tutor actually changed the move. Check whether the learner received the intended opportunity. Read what the learner did next. Repeat under another relevant condition. Adapt or reject the coaching hypothesis when evidence disagrees. Retire the target when the tutor can carry it independently.

Do not call agreement a coaching receipt. Do not call reflection a coaching receipt. Do not call a rehearsal a coaching receipt. Do not call one observer-visible performance a stable coaching receipt.

The Coaching Receipt is changed teaching that survives the return to live complexity and creates a learner opportunity that did not reliably exist before.

That is Tutor Handbook Volume 0075.

Small Group Tutorials

The Tutor Handbook Vol No.0075 | The Coaching Receipt — How a Tutor Shows That Professional Feedback Changed Live Teaching Rather Than Only Producing Agreement