The Tutor Handbook · Volume 0203 · Series ID THB-0203
The Tutor Handbook: Complete Series Index
When the same remedy is prescribed for every implementation problem
A tuition programme introduces a teaching routine that appears sensible on paper. Tutors are meant to use a brief retrieval check at the start of selected lessons, read the response, and decide whether to continue, repair or branch. After six weeks, use is patchy.
The obvious response is to train tutors again.
That may be the right response. It may also be exactly the wrong one.
One tutor understands the routine but has no time because the opening five minutes are already occupied by attendance, homework collection and school-paper triage. Another tutor has time but does not know how to interpret the retrieval evidence. A third tutor understands both the routine and the evidence but is using an old worksheet that does not contain the required prompts. A fourth tutor thinks the routine is educationally sound but sees no reason to use it with a learner whose current problem is already obvious. A fifth tutor uses it faithfully, but the learner arrives late often enough that the opening check is the first thing dropped.
These are not five versions of the same training problem.
They are five different implementation barriers.
The Implementation-Barrier Response Gate asks a programme to identify what is actually stopping an intended practice before choosing the implementation response. Its central discipline is simple: do not prescribe training for a workflow problem, motivation for an access problem, monitoring for a material problem, or stricter fidelity for a practice that does not fit the case.
Implementation improves when the response is matched to the barrier.
Quick answer
Before changing the tutor, the material, the timetable or the practice, name the barrier at the level where it operates.
If the tutor does not know what to do, knowledge-building or modelling may be needed.
If the tutor knows what to do but cannot execute it fluently, rehearsal, coaching or live feedback may be needed.
If the tutor can do it but ordinary lesson conditions make it impractical, redesign the workflow, timing, materials or scope.
If the tutor sees the practice as inappropriate or unreasonable, investigate acceptability rather than labelling the reaction resistance.
If only some groups receive the practice, investigate reach and eligibility.
If the practice is delivered but the active ingredient has disappeared, investigate fidelity and adaptation.
If the barrier comes from the surrounding programme—conflicting systems, duplicate documentation, unclear ownership, inaccessible tools, unstable schedules—fix the system rather than repeatedly retraining the person inside it.
The aim is not to find one cause with certainty. It is to reduce ambiguity enough to choose the smallest response that could plausibly remove the barrier, then check whether implementation changes.
A good implementation response behaves like a diagnostic intervention: it should be specific enough that its success or failure teaches the programme something.
The ownership boundary
This gate sits beside several existing Tutor Handbook owners.
The Training Need Gate asks whether a tutor difficulty is genuinely a knowledge, skill or judgement gap before more training is assigned. That is the tutor-development differential.
The Implementation-Feasibility Gate asks whether a competent ordinary tutor can carry the practice inside real time, materials and workflow constraints.
The Implementation-Acceptability Gate asks how tutors, learners and families experience and judge a practice without turning popularity into effectiveness.
The Implementation-Readiness Gate asks whether the programme is ready to launch at all.
The present article owns the next decision: once a barrier is observed, what kind of response belongs to that barrier?
That sounds administrative until the wrong response begins consuming learner time.
A programme that repeatedly solves the wrong problem creates implementation theatre. More workshops are held. More checklists are created. More reminders are sent. Tutors appear busier. The original practice remains fragile.
Current implementation guidance treats barriers and responses as a matching problem
The Australian Education Research Organisation’s practice guide Identifying and addressing implementation enablers and barriers, published 31 March 2025 and updated 15 September 2026, explicitly frames implementation as a process of identifying factors that help or hinder a practice, prioritising them, responding and monitoring what happens next. It is research-informed school implementation guidance, not a trial of this Tutor Handbook gate.
AERO’s Developing a school implementation plan, published 31 March 2025 and updated 29 April 2026, similarly separates the work of identifying enablers and barriers from selecting implementation strategies. The sequence matters: strategy follows diagnosis.
AERO’s Context first: Tailoring implementation to each school setting, published 15 September 2026, emphasises that implementation should be adapted to local context rather than treated as one-size-fits-all.
The Education Endowment Foundation’s A School’s Guide to Implementation, third edition published 24 April 2024, also treats implementation as a process requiring attention to context, infrastructure, engagement and reflection rather than a one-off launch.
These sources justify the discipline of diagnosis before response. They do not provide a universal private-tuition algorithm in which each barrier maps mechanically to one remedy. The mapping remains a professional judgement that should be tested against local evidence.
Start with the observable failure, not the preferred explanation
A weak implementation conversation often begins with a cause.
“Tutors are not bought in.”
“They need more training.”
“Students are resistant.”
“The material is too hard.”
“The process is inconsistent.”
These may be true. They are not observations.
Begin one step earlier.
What is actually happening?
The retrieval check is absent in six of ten eligible lessons.
The check is present, but tutors look at the answer and continue with the planned worksheet regardless.
The practice appears with experienced tutors and disappears with new tutors.
The routine works in Mathematics and is rarely used in English.
The routine is used when sessions begin on time and skipped after late arrivals.
The record is completed after the lesson rather than while the decision is still live.
The learner participates when the response is private and withdraws when the same response is public.
These statements do not yet explain the cause. That is their strength. They keep several explanations alive long enough for a discriminating check.
A useful barrier map
For ordinary tuition, barriers can be organised into seven practical families. The families are not diagnoses. They are starting points for choosing what to inspect next.
Knowledge barrier. The tutor does not understand the practice, its purpose or the conditions under which it should be used.
Skill barrier. The tutor understands the idea but cannot yet perform the move reliably in live teaching.
Judgement barrier. The tutor can perform the move but has difficulty deciding when, why or with whom to use it.
Workflow barrier. The practice competes with time, sequence, documentation, materials, technology or other lesson operations.
Access or participation barrier. A learner, tutor or family cannot participate fairly because of language, modality, technology, schedule, privacy or another legitimate condition.
Acceptability or value barrier. People understand the practice but regard some part of it as inappropriate, disproportionate or misaligned with an important value.
System-coherence barrier. Other programme rules, school methods, reporting requirements or ownership arrangements push behaviour in the opposite direction.
A real case can contain more than one family. The point is not to force it into a box. The point is to prevent the first convenient story from becoming the intervention.
Composite case: the training that makes no difference
This case is fictional and constructed for teaching.
A programme wants tutors to use one fresh independent question after correcting a misconception. The fresh question is the receipt that separates copied correction from actual repair.
Use is weak. Leaders schedule a refresher session.
Tutors can already explain the rationale accurately. In rehearsal, they can run the routine. During live lessons, however, the fresh question is often omitted.
The programme initially concludes that transfer from training to practice is weak.
A coach observes four sessions. The problem appears in the final fifteen minutes. Tutors spend that period marking homework, setting next work, answering parent messages and trying to finish the planned worksheet. The fresh check competes with several closing tasks.
The barrier is not primarily knowledge or skill.
The programme changes the response. It removes a duplicate end-of-lesson note, moves homework collection earlier, and includes one ready-to-use fresh item in the lesson material. The coach keeps a light sample to see whether the routine now appears.
If use improves, the programme has evidence that workflow was load-bearing.
If use does not improve, the diagnosis remains open.
The important move was not “less training”. It was refusing to use training as a default explanation for non-use.
Match knowledge barriers with knowledge-building
When tutors genuinely do not understand the practice, training is appropriate.
But training should teach the missing object.
If tutors cannot explain why a private first response matters, teach the evidence problem it solves.
If tutors do not recognise the active ingredient, model strong and weak examples.
If tutors know the steps but not the eligibility condition, compare cases in which the practice should and should not be used.
Knowledge-building should end with a check that can fail.
Can the tutor identify the practice in a new example?
Can they explain why one adaptation is safe and another removes the mechanism?
Can they choose the right case without being told the answer?
Attendance at a workshop is not the receipt.
The Professional-Learning Cycle Gate provides the larger owner for moving from knowledge-building through modelling, rehearsal, live practice, coaching and review.
The barrier gate decides whether that cycle is the right response in the first place.
Match skill barriers with rehearsal and feedback
Some tutors understand a teaching move but cannot execute it fluently.
They know that a prompt should preserve learner thinking. In live teaching, their prompt becomes a mini-solution.
They understand that feedback should point to the next action. Under time pressure, they give three corrections at once.
They know how to ask a diagnostic question. When the learner hesitates, they rephrase so heavily that the question no longer discriminates between causes.
These are performance problems.
More explanation can produce diminishing returns. The tutor may need to see the move modelled, rehearse a small segment, receive specific feedback and try again.
Keep the rehearsal close to the real demand. If the problem appears when three learners respond differently, a one-to-one scripted role-play may be too clean. If the problem appears under examination-season time pressure, practise the compressed version rather than only the ideal version.
A skill response should change observable execution.
If it does not, reconsider the barrier.
Match judgement barriers with cases, contrasts and decision rules
Judgement is different from skill.
A tutor may know how to fade a scaffold and still not know when to fade it.
They may be excellent at explaining but over-explain because they cannot tell whether the learner needs another model or an independent attempt.
They may run a formative check well but use it on every task, including low-value moments where no decision depends on the result.
Judgement barriers need decision practice.
Use contrastive cases.
“Same learner, different evidence. Which route now?”
“Same practice, different task purpose. Is it eligible?”
“Same result, different support condition. Does the conclusion change?”
Ask the tutor to commit before the answer is revealed.
Then discuss the evidence feature that should change the decision.
This builds conditional knowledge rather than another procedure.
Match workflow barriers with redesign, not exhortation
A tutor can believe in a practice, know how to use it and still fail to deliver it because the environment makes non-use the easiest path.
The material is stored in a separate system.
The login expires.
The worksheet requires printing.
The evidence form duplicates the lesson note.
The routine occurs exactly when parents arrive for handover.
The practice assumes ten uninterrupted minutes that rarely exist.
These are implementation design problems.
The Implementation-Feasibility Gate asks whether the practice fits ordinary work. The barrier response is to change the surrounding workflow while preserving the active ingredient.
Put the resource where tutors already work.
Remove duplicate fields.
Pre-build the fresh question.
Shift the routine earlier.
Narrow the practice to cases where the value justifies the time.
A workflow fix should reduce friction without quietly removing the educational job.
Match access barriers with access repair
A practice can fail because an intended participant cannot use it fairly.
An online whiteboard is inaccessible on the learner’s device.
A written reflection assumes a language-production level unrelated to the target.
A private digital response requires an account the learner does not have.
A parent communication system assumes a particular language or schedule.
Do not interpret these as motivation.
The response is access repair.
Change the route while preserving the construct. Provide the legitimate support. Offer another response format where the medium is not the target. Move communication to a usable channel.
Then check reach again.
The Response-Modality Equivalence Gate and Accommodation Baseline remain relevant when response or access conditions change.
Match acceptability barriers with mechanism-level inquiry
A disliked practice can still be educationally useful.
A liked practice can still be weak.
That is why the response to low acceptability should not be immediate abandonment.
Ask what exactly people object to.
Is it public exposure?
Extra unpaid work?
Loss of tutor discretion?
A confusing rationale?
A real value conflict?
A mismatch with one subject?
A routine that feels punitive?
A practice that is fine but introduced badly?
The Implementation-Acceptability Gate owns the interpretation. The barrier response may be communication, surface redesign, narrower eligibility, stronger privacy, or a decision to stop.
The mistake is to call all negative reaction resistance and then prescribe persuasion.
Match coherence barriers with owner reconciliation
Sometimes the tutor is caught between two legitimate systems.
The tuition programme asks for one notation. The school uses another.
The programme says “delay feedback until the reattempt”. The homework deadline requires immediate correction.
A coach asks for detailed notes. The evidence-capture standard says documentation should be minimal.
One system rewards consistency; another rewards adaptation.
If tutors receive conflicting instructions, inconsistency is predictable.
The response is not more reminders from both sides.
Someone with legitimate authority must reconcile the conflict.
The Curriculum-Method Reconciliation Gate owns school-versus-tuition method coherence. The Escalation-Authority Gate helps decide when the tutor should move the decision upward.
A coherent system should not require tutors to solve policy contradictions privately in the lesson.
The barrier can move after the first fix
Implementation diagnosis is dynamic.
A programme removes the workflow barrier. Use increases. Now quality varies because tutors need more judgement practice.
A programme provides training. Tutors become competent. Now learners object to the public response format.
A programme changes the format. Acceptability improves. Now the practice takes too long in examination season.
The barrier has moved.
Do not treat this as failure of the original diagnosis. Complex systems often reveal the next constraint only after the first one is removed.
The programme should keep the intervention history visible.
“What changed?”
“What improved?”
“What did not?”
“What is now the limiting factor?”
This produces a sequence of bounded improvements rather than one large reform story.
Composite case: the new tutor who appears “unmotivated”
This case is fictional.
A newly recruited tutor rarely uses the programme’s error-classification routine. The programme lead is concerned because the tutor completes other requirements conscientiously.
A quick conversation reveals that the tutor believes the classification must be recorded for every error. The documentation would take several minutes per lesson.
The intended practice was narrower: classify only recurrent or decision-relevant errors when the category changes the next instructional move.
The barrier is partly knowledge and partly design communication.
The tutor did not reject the educational idea. They inferred a much larger workload than the programme intended.
The response is a corrected example showing when no classification is needed, plus a simplified note template.
The next observation checks whether the tutor now uses the routine selectively.
Calling this “motivation” would have produced the wrong relationship problem and the wrong implementation response.
Do not infer barriers from personality
Implementation language can become a disguised character judgement.
“Not reflective.”
“Not innovative.”
“Old school.”
“Doesn’t care about data.”
“Too rigid.”
These labels are low-resolution.
A tutor may look rigid because the materials are tightly scripted.
They may avoid data because the dashboard is slow and duplicative.
They may reject a new routine because it conflicts with a valid subject-specific demand.
They may genuinely lack a needed skill.
The programme should test the instructional explanation before turning implementation into identity.
This matters ethically and practically. Character explanations are difficult to falsify. Operational explanations can be checked.
A small experiment is often better than a large implementation response
Suppose the programme suspects that tutors skip a routine because it takes too long.
Do not redesign the entire handbook first.
Choose a bounded test.
For one week, pre-load the required material and remove one duplicate record. Compare whether use changes in eligible lessons.
If use rises, time and friction gain support as explanations.
If use does not move, the workflow hypothesis weakens.
The experiment does not need statistical power for a causal claim. It is an implementation probe designed to improve the next decision.
Keep the conclusion proportionate.
“We removed one source of friction and use increased in this setting.”
Not:
“We proved workload caused non-use.”
Implementation strategies should have their own receipt
A strategy is another intervention.
Training can fail.
A new template can fail.
A reminder can fail.
A coaching cycle can fail.
A timetable change can create a new problem.
Therefore every material implementation response should have a receipt.
If the barrier was knowledge, can tutors now explain and discriminate the practice?
If the barrier was skill, does live execution improve?
If the barrier was workflow, does use increase under ordinary conditions?
If the barrier was acceptability, does the specific concern change after redesign?
If the barrier was access, do previously excluded users now receive the practice?
If the barrier was coherence, do tutors still receive conflicting signals?
Without a receipt, programmes accumulate implementation strategies in the same way they accumulate teaching routines.
Avoid the monitoring reflex
When leaders see inconsistent practice, they often increase monitoring.
More observations.
More checklists.
More compliance reports.
Monitoring can clarify what is happening. It does not automatically remove the cause.
If tutors lack knowledge, surveillance does not create knowledge.
If the material is inaccessible, a checklist does not create access.
If the workflow is overloaded, another reporting requirement makes the barrier worse.
If the practice is unacceptable because it exposes learners publicly, observing it more often does not repair dignity.
Use monitoring to reduce uncertainty, then act on the barrier it reveals.
The Tutor-Observation Sample Gate remains the owner for choosing representative observation samples without turning coaching into surveillance.
A practical barrier-response protocol
When an implementation problem appears, write the problem in observable form.
“Only four of nine eligible lessons used the fresh verification question.”
List the strongest plausible barrier families.
Knowledge? Skill? Judgement? Workflow? Access? Acceptability? Coherence?
Choose the smallest discriminating check.
Ask the tutor to explain the purpose. Observe one live case. Inspect the material path. Time the routine. Ask the learner about the response condition. Compare a lesson where the practice appeared with one where it did not.
Then choose the smallest response that directly addresses the best-supported barrier.
Do not change five things at once.
State what should happen if the response is correct.
“After the template is integrated into the normal lesson page, eligible use should rise without additional training.”
Review the receipt.
If the expected change appears, keep monitoring lightly.
If it does not, update the barrier model.
The method is diagnostic, not punitive.
Failure modes
The training reflex. Every implementation problem produces another workshop.
The motivation story. Non-use is attributed to attitude before time, skill, access, fit and system contradictions are checked.
The monitoring reflex. More observation is treated as a treatment rather than an information source.
The strategy pile. Training, reminders, templates and coaching are added together, making it impossible to know what removed the barrier.
The single-cause fantasy. A complex pattern is forced into one explanation even when several constraints interact.
The personality diagnosis. Tutors are labelled resistant, rigid or careless instead of the work system being inspected.
The fidelity-first failure. Tutors are told to comply more closely even when the practice is infeasible or mismatched.
The surface-fix failure. Friction is removed by deleting the active ingredient.
The no-receipt response. An implementation strategy is launched but nobody defines what observable change would count as success.
The permanent-response failure. Temporary launch support becomes a permanent burden because the programme never checks whether the barrier still exists.
Barrier evidence should travel with the response
Once a programme chooses an implementation response, the original barrier evidence should not disappear.
Suppose tutors are not using a new feedback routine because the lesson workflow leaves no time for the required learner reattempt. The programme removes a duplicate recording step and protects five minutes for the reattempt.
The later review should not ask only, “Are tutors using the routine now?”
It should also ask whether the identified barrier actually changed.
Is the reattempt now happening within ordinary lesson time?
Did duplicate administration fall?
Are tutors still reporting the same bottleneck?
Did another barrier appear once the first one was removed?
This matters because implementation can improve for reasons unrelated to the chosen response. A motivated tutor may compensate temporarily. A lighter examination week may create spare time. One enthusiastic subject team may carry the routine while others still cannot.
Keeping the barrier and response connected creates a more interpretable receipt.
The programme can say:
“We believed the main barrier was duplicated workflow. We removed the duplication. Ordinary completion improved and tutors no longer report the same time conflict.”
That is still not causal proof in a research sense.
It is a stronger implementation story than:
“We reminded everyone again and usage went up.”
One barrier can require several owners
Some implementation problems cross levels.
A tutor may need skill rehearsal.
The programme may need a better material.
The timetable may need a small change.
The learner may need a legitimate access adaptation.
The family may need a clearer explanation of the practice.
Trying to force one owner to solve all of these creates frustration.
The tutor gets more training for a scheduling problem.
The programme lead redesigns materials for a learner-specific access need.
The family is asked to motivate the child when the task is simply too hard to enter independently.
A useful response plan therefore names the smallest legitimate owner for each part of the barrier.
Tutor-owned: practise the teaching move.
Coach-owned: observe and give bounded feedback.
Programme-owned: remove duplicate workflow.
Family-owned: provide a practical home condition where appropriate.
External owner: resolve a school or access decision outside the tutor’s authority.
The implementation response becomes a coordinated set of bounded actions rather than a vague instruction to “improve fidelity”.
That division also protects accountability.
People should be held responsible for the part they can actually change.
Evidence boundaries
AERO’s Identifying and addressing implementation enablers and barriers, published 31 March 2025 and updated 15 September 2026, is research-informed school implementation guidance. It supports systematic barrier identification and response, but it does not validate the seven barrier families used in this private-tuition framework.
AERO’s Developing a school implementation plan, published 31 March 2025 and updated 29 April 2026, separates barrier identification, strategy selection, outcome definition and adjustment. This supports the sequencing principle used here.
AERO’s Context first, published 15 September 2026, is a practice guide using school examples to illustrate contextual tailoring. It should not be treated as evidence that one contextual adaptation works everywhere.
EEF’s A School’s Guide to Implementation, third edition published 24 April 2024, is implementation guidance based on research and practitioner expertise. It supports structured, contextual, socially informed implementation rather than one-off delivery.
The local barrier-response map in this article remains a professional decision aid. Its quality depends on whether it helps the programme choose better, testable responses without overclaiming causes.
The end state
A good programme does not ask, “Why are tutors not doing what we told them?”
It asks a more useful question.
“What is making the intended practice hard, unclear, inappropriate or impossible here—and what is the smallest response that would remove that obstacle without removing the educational job?”
Sometimes the answer is training.
Sometimes it is rehearsal.
Sometimes it is one clearer example.
Sometimes it is a shorter form.
Sometimes it is a timetable change.
Sometimes it is an access repair.
Sometimes it is a conversation about values.
Sometimes the right answer is to stop or narrow the practice.
The Implementation-Barrier Response Gate exists to keep those possibilities open long enough for evidence to choose among them.
Implementation quality improves when the programme stops prescribing its favourite remedy and starts matching the response to the barrier that is actually in the way.
Prioritise the barrier that changes the decision
A programme can identify several real barriers at once and still make a poor response by treating them as equally urgent.
Suppose tutors report three problems with a new formative routine. The digital form is mildly irritating. New tutors are unsure how to interpret one response pattern. One group of learners cannot access the form at all on the device they use.
All three observations matter. They do not deserve the same first action.
Prioritisation should follow educational consequence, reach and reversibility.
A barrier deserves earlier attention when it prevents the intended learner from receiving the practice, corrupts the evidence on which later decisions depend, creates a safety or dignity concern, or makes the practice impossible under normal conditions. A low-cost irritation can wait if the educational job still works. An access barrier should not wait merely because it affects fewer people.
The same logic protects the programme from chasing the loudest complaint. The most visible barrier is not always the most consequential one.
Ask:
What decision is currently being damaged?
How many eligible cases are affected?
What is the cost of waiting?
Can the suspected barrier be tested with a small reversible change?
Would fixing this barrier expose the next constraint, or is another barrier upstream?
This creates an implementation order.
For example, repair access before measuring acceptability among people who could not properly participate. Clarify the active ingredient before judging fidelity. Resolve a system contradiction before coaching tutors to execute both conflicting expectations better. Restore basic feasibility before interpreting non-use as a value problem.
Barrier work becomes far more efficient when the programme removes upstream constraints first.
That is also why the gate should resist a long implementation checklist. The purpose is not to score every possible barrier. It is to find the one or two constraints that currently change the educational decision, respond proportionately, and learn from the result.
