The Tutor Handbook Vol No.0201 | The Implementation-Readiness Gate — How a Tuition Programme Decides Whether It Is Actually Ready to Launch an Evidence-Informed Teaching Practice Before Training Everyone or Blaming Tutors for a Predictable Rollout Failure

The Tutor Handbook · Volume 0201 · Series ID THB-0201 The Tutor Handbook: Complete Series Index ## A good practice can fail because the programme began before it was ready A tuition programme finds a teaching practice worth using. The evidence is promising. The educational problem is real. The practice has a plausible mechanism. Leaders are enthusiastic. Tutors are told that the change will begin next Monday. Then ordinary reality arrives. The training material is not finished. Two subject teams interpret the practice differently. The worksheet needed to support the routine exists only for Primary Mathematics. Nobody has decided who owns updates. Tutors do not know whether the practice is mandatory in every lesson or only under particular conditions. There is no plan for what evidence will show whether the change is working. One tutor raises a privacy concern about the digital tool. Another points out that the new routine adds ten minutes to a ninety-minute lesson whose existing plan is already full. After a month, implementation is uneven. The programme says tutors need to be more consistent. Perhaps. But the more important question may be: was the programme ready to ask for consistency? The Implementation-Readiness Gate asks whether a tuition programme has enough clarity, capacity, ownership, materials, professional learning, evidence design and contextual fit to begin an implementation effort responsibly. It sits before broad rollout. Its purpose is to prevent predictable launch problems from being mislabelled later as tutor resistance, weak fidelity or lack of commitment. Readiness is not perfection. No programme can remove every uncertainty before action. The gate asks whether the uncertainties that remain are safe to learn from rather than obvious reasons the rollout will be unreadable or unfair. ## Quick answer Before launching an evidence-informed practice programme-wide, check whether seven conditions are sufficiently ready. Problem: the programme can state the educational problem the practice is meant to solve and why it matters now. Practice: tutors can identify the active ingredient, the conditions where the practice belongs and the surface features that may adapt. Context: the practice has been checked against real group size, subjects, session length, learner access needs, school alignment and current workload. Capacity: the programme has enough time, materials, technology, coaching and ownership to support ordinary use. People: tutors know what they are being asked to learn, have a route to practise it, and can raise implementation concerns without being treated as obstructive. Evidence: the programme knows what early signals will reveal feasibility, fidelity, reach, acceptability and learner opportunity without turning the launch into continuous surveillance. Decision route: someone has authority to adapt, pause, narrow, continue or stop the effort when evidence arrives. If several of these are materially absent, do not pretend a full rollout is a fair test. Prepare first, or run a bounded trial whose purpose is explicitly to learn about readiness. ## Readiness is different from whether the practice itself is feasible The Implementation-Feasibility Gate asks whether a competent ordinary tutor can realistically carry a teaching routine inside real time, material, skill and workflow constraints. The readiness question is one level earlier and wider. A practice may be feasible in principle while the programme is not ready to implement it. For example, a two-minute exit-ticket routine may fit lessons perfectly. But if tutors have no shared definition of the learning target, the programme has no usable item bank, no one owns the results and leaders have not decided what tutors should do when the exit ticket exposes a gap, the programme is not ready for a useful rollout. The opposite can also occur. A programme can be beautifully prepared to implement a routine that is intrinsically too burdensome. Readiness does not rescue a bad fit. Treat the two gates separately. “Can this practice work here?” and “Are we prepared to begin?” are different questions. ## Current implementation guidance makes readiness an explicit job The Australian Education Research Organisation published How ready are we? Preparing to start an implementation effort on 15 September 2026. AERO describes the resource as helping school leaders reflect on preparedness to begin an implementation effort and decide next steps. It emphasises deliberate planning, adequate resourcing, ongoing commitment, willingness and capacity. That is research-informed school implementation guidance. It is not a validated readiness score for private tuition. The useful principle is that beginning implementation is itself a decision that deserves evidence and preparation. EEF’s A School’s Guide to Implementation, third edition published 24 April 2024, similarly separates Explore and Prepare from Deliver and Sustain. It encourages leaders to consider suitability and feasibility before moving into implementation. AERO’s Context first: Tailoring implementation to each school setting, published 15 September 2026, stresses that implementation is not one-size-fits-all and should respond to local context and needs. A tuition centre is not a school system. It is often smaller, faster and closer to individual learners. That can make implementation easier. It can also make hidden dependencies more dangerous because one person may hold the materials, training, judgement and technical knowledge. Readiness matters precisely because small systems can move quickly. ## Start by proving there is a problem worth changing Implementation enthusiasm can create solutions looking for problems. A programme sees a compelling practice guide, conference presentation or tool and wants to adopt it. The practice may be excellent. If the local problem is weak or undefined, implementation becomes difficult to interpret. Suppose leaders want every tutor to introduce structured retrieval at the start of lessons. What problem is this meant to solve? Learners forgetting prior knowledge? Tutors spending too much time reteaching? Weak cumulative review? Poor examination retrieval? A generic belief that retrieval is “evidence based” is not yet a local implementation problem. The programme should identify enough local evidence to justify action. That might be repeated loss of prerequisites, learner work showing strong immediate practice but weak delayed access, or tutors independently reporting that earlier content repeatedly collapses. The evidence does not need to be a formal research study. It needs to be strong enough to show why scarce tutor attention should move here instead of somewhere else. Readiness begins with a problem statement that can later be revisited. ## Define what the practice actually is Many rollout failures begin with a name that everybody thinks they understand. “Explicit instruction.” “Formative assessment.” “Retrieval practice.” “Differentiation.” “AI-supported tutoring.” The labels are too broad to implement. Before rollout, write the smallest operational definition tutors need. What is the active ingredient? What learner opportunity should it create? What are two or three observable features? What can vary? When should a tutor not use it? For a feedback-action loop, the active ingredient might be that feedback creates a learner action on the target and later evidence shows whether the learner can use the feedback beyond the corrected item. The exact wording, colour of annotations and task length can vary. The Active-Ingredient Adaptation Gate protects this distinction after implementation begins. Readiness requires it before tutors are asked to adapt. If leaders cannot state what must remain true, tutors cannot be held responsible for preserving it. ## Check the local context before standardising the surface A practice that fits one subject, age or delivery mode may need a different surface elsewhere. Primary 3 Science, Secondary 4 English and JC Mathematics do not create identical instructional conditions. A three-learner in-person class differs from a one-to-one online session. A ninety-minute lesson differs from an intensive examination clinic. A learner using legitimate access support may need different response conditions from peers. Context inspection should not become an excuse to reject every common standard. Ask which features of context could plausibly change implementation. Session time. Group size. Task type. Curriculum language. Tutor expertise. Learner age. Technology access. Privacy. Home requirements. School method alignment. Existing routines competing for the same minutes. Now decide what should be common and what may be tailored. This protects the programme from two mistakes: forcing one rigid form everywhere and allowing every tutor to invent a different practice under the same name. ## Composite case: the exit ticket that had nowhere to go The following case is fictional and constructed for teaching. A programme decides that every lesson should end with an exit ticket because tutors need better evidence of learning. The idea is sensible. Tutors are briefed. A folder of questions is created. The routine begins on Monday. By Friday, implementation looks inconsistent. Some tutors use one question. Others use three. Some mark the ticket during the lesson. Others send it as homework. One tutor records scores in a spreadsheet. Another simply glances at the answer. A third skips the ticket because the lesson ran long. Leaders initially see a fidelity problem. A readiness review reveals something earlier. Nobody had defined the decision the exit ticket should support. Was it for planning the next lesson? Checking independent performance? Reporting to parents? Comparing tutors? The item bank mixed these purposes. Tutors had no protected time to review results and no rule for what to do when one learner failed while two succeeded. The programme pauses universal use. It redefines the practice: one short independent item aligned to the session’s key learning target, reviewed before the next lesson, used to decide whether the next session begins with advance, brief repair or a discriminating check. Results are not used for tutor ranking or parent reporting. Now the practice has a job. The original problem was not that tutors refused an evidence-informed routine. The programme launched a visible action before designing the decision it was supposed to feed. ## Capacity means more than available minutes Readiness capacity has several forms. Time capacity: can tutors learn, prepare and execute the practice without displacing a more important job? Material capacity: are examples, tasks, answer guidance or access versions available at sufficient quality? Skill capacity: do tutors already possess the prerequisite professional capabilities, or is a professional-learning cycle needed first? Coaching capacity: who will help when early implementation becomes ambiguous? Technical capacity: if a tool is involved, do accounts, devices, privacy boundaries and failure routes exist? Decision capacity: is someone actually able to interpret early evidence and authorise changes? A programme can have money and still lack capacity. It can have expert tutors and still lack shared materials. It can have excellent materials and still lack protected review time. Readiness makes the hidden infrastructure visible before tutors are blamed for improvising around it. ## Ownership should be named before launch “Everyone owns implementation” is not enough. Who owns the current practice definition? Who updates materials? Who answers tutor questions? Who decides whether an adaptation is legitimate? Who reviews early implementation evidence? Who can pause the rollout? Who communicates significant changes? In a small tuition programme, one person may hold several roles. That is fine if the roles are explicit. The danger is orphaned responsibility. A tutor notices that the shared worksheet no longer matches the current syllabus. They assume someone else will update it. A coach sees that tutors are interpreting the same routine differently but has no authority to change the documentation. Leaders assume coaches have handled the problem. Readiness requires a short ownership map, not an organisational chart. The goal is to prevent implementation questions from circulating without a legitimate decision-maker. ## Tutor willingness is not the same as tutor readiness A tutor can be enthusiastic and unprepared. A tutor can be sceptical and highly capable. Do not confuse attitude with readiness. Ask what the tutor needs to know and do. Can they explain the problem? Can they recognise when the practice applies? Can they execute its core move in rehearsal? Do they understand the access boundary? Do they know what evidence to notice? Do they know where to raise a problem? If not, build the professional-learning route. The Professional-Learning Cycle Gate owns that route once this article is published in sequence. Readiness should also make room for tutor concerns. A tutor who says, “This will add duplicate note-taking” may be identifying a feasibility problem. A tutor who says, “This example will confuse the school method” may be identifying a coherence issue. A tutor who says, “I prefer my old approach” may be expressing a preference that does not outweigh stronger evidence. The programme needs enough inquiry to tell the difference. ## Learner and family conditions can determine launch readiness Some practices require no family involvement. Others change homework, technology use, scheduling or data sharing. If the implementation changes what families must do, the programme should inspect that condition before launch. Does the learner need a device at home? Will a parent receive more messages? Does homework now require printing? Will recorded audio or video be stored? Does the practice alter the learner’s experience of support or independence in a way that needs explanation? The Implementation-Acceptability Gate asks how stakeholder reactions should be interpreted after and during implementation. Readiness asks whether predictable participation conditions have been considered before people are asked to live with the practice. A programme does not need unanimous enthusiasm. It needs to remove obvious preventable barriers and explain material changes honestly. ## Define the minimum viable evidence before rollout Implementation can drown in measurement. Leaders want to know whether tutors used the practice, whether learners benefited, whether parents liked it, whether the materials worked and whether the programme should continue. They build a form with twenty fields. Tutor burden rises before the practice has a chance to stabilise. Readiness should define the minimum evidence needed for early decisions. For example: one observation of the target practice during an ordinary lesson; a short tutor note on the main friction; a learner work sample showing the intended opportunity; a count of groups where the practice was actually used when eligible; a targeted acceptability question if participation is a known risk. The evidence should map to decisions. If no one knows what would change when a field is completed, remove the field. The Evidence-Capture Burden Gate applies from day one. ## Predefine what early trouble will mean Every new practice has roughness. The programme should not interpret the first problem after it occurs without a framework. Before launch, name plausible failure classes. If tutors understand the practice but cannot fit it into ordinary lesson time, suspect feasibility. If they use a visibly different routine, inspect fidelity and clarity. If some eligible groups never receive it, inspect reach. If learners or tutors experience avoidable friction, inspect acceptability. If the practice is used correctly but the intended learner opportunity does not appear, inspect the practice-task fit and active mechanism. If only one expert tutor can make it work, inspect skill requirements and sustainability. This does not predict every failure. It gives leaders a vocabulary that prevents “implementation failed” from becoming one undifferentiated conclusion. ## Decide whether to pilot or launch Not every uncertainty requires a pilot. If the practice is low-risk, simple, reversible and well understood, a small programme may implement broadly while monitoring lightly. A bounded pilot becomes more useful when the practice is complex, costly, technically dependent, privacy-sensitive, difficult to reverse or uncertain in the local context. The purpose of a pilot should be declared. Are we testing learner usefulness? Tutor feasibility? Material quality? Technology reliability? Adaptation across subjects? Professional-learning demands? A pilot without a question becomes a small rollout whose results are hard to interpret. The Trial Run owns learner-route trials. This readiness gate applies the same disciplined idea at programme-practice level: bounded use should reduce a known uncertainty before expansion. ## Composite case: the AI question generator that was technically ready and educationally unready This case is fictional. A tuition programme buys access to an AI tool that can generate practice questions rapidly. The technical launch is smooth. Accounts work. Tutors can enter prompts. The tool produces attractive worksheets. Within two weeks, problems appear. Question difficulty is inconsistent. Some answer keys are wrong. Tutors use different prompt styles. One tutor sends generated questions directly to learners; another manually verifies every item. Parents assume the tool is choosing the child’s learning route. The programme had technical readiness and lacked educational readiness. A reset defines the job. AI may draft questions after the tutor declares the target, difficulty, representation and intended misconception. A human tutor verifies every item and answer before learner use. Generated practice is not accepted as evidence of mastery unless the task conditions are appropriate. The tool does not decide diagnosis or progression. The programme creates two tested prompt patterns, a short verification checklist and a route for reporting recurring failures. Tutors rehearse on deliberately flawed outputs before live use. The technology did not become safer because tutors were told to “be careful”. The implementation became ready because ownership, boundaries, evidence and professional skill were designed. ## Readiness includes a rollback or pause route Leaders often worry that announcing a pause will make implementation look weak. The opposite can be true. A responsible programme should know how to reduce scope when evidence says the launch is not ready. Pause new groups while supporting existing ones. Return to a previous stable routine. Limit the practice to one subject while materials are improved elsewhere. Remove a technical feature while preserving the educational mechanism. Continue professional learning before expanding. A pause is not automatically failure. It can be evidence that the programme is capable of learning before harm, burden or confusion spreads. The important distinction is between reversible learning and chaotic retreat. Name the condition that would justify pausing and the evidence needed to restart. ## Readiness should include accessibility from the beginning Accessibility cannot be an afterthought added after the standard version has already become normal. If the practice uses a response format, ask whether legitimate alternative response routes are needed. If it uses printed materials, check readability and whether digital or enlarged versions are required. If it uses audio, visual or digital platforms, check the access implications before assuming every learner can participate equivalently. If the practice is intended to test independence, distinguish access support from answer-giving help. The Response-Modality Equivalence Gate and the Access-Support Boundary protect these decisions after launch. Readiness means the programme has anticipated the common cases rather than leaving each tutor to invent an accessibility policy in the moment. ## Do not launch because the calendar says it is time Programmes like dates. “New term, new system.” “Start in January.” “Roll out after the staff meeting.” A calendar can coordinate work. It should not overrule material readiness. If the active ingredient is still unclear, the materials are unstable and the coaching route does not exist, a launch date does not make those conditions disappear. At the same time, perfectionism can prevent useful change forever. Use a readiness threshold rather than an ideal state. Which conditions must be true before the learner-facing rollout is responsible? Which uncertainties can be learned during implementation? Which unresolved issues are high consequence enough to delay? This turns “Are we ready?” into a decision rather than a feeling. ## A simple readiness review for a small tuition programme The review can fit on one page. Problem: What local evidence says this problem deserves attention now? Practice: What is the active ingredient and where does it not belong? Context: Which subjects, ages, group structures or access conditions require adaptation? Capacity: What time, materials, technology and professional skill are required? Ownership: Who maintains the practice and who can make implementation decisions? Professional learning: What must tutors know, rehearse and demonstrate before ordinary use? Evidence: What minimum signals will tell us whether implementation is feasible, coherent and reaching the intended groups? Voice: Whose experience could reveal a burden or barrier leaders cannot see? Decision: What would make us continue, adapt, narrow, pause or stop? Then classify the result. Ready for broad implementation. Ready for a bounded pilot. Ready after one specific preparation gap is closed. Not ready because the problem or practice is still too unclear. The classification should create action, not paperwork. ## What not-ready should trigger “Not ready” should not become an indefinite holding category. If the missing condition is materials, assign an owner and deadline. If it is tutor skill, run the professional-learning cycle. If it is unclear context fit, test the practice in a bounded representative setting. If it is privacy or access, resolve the boundary before learner use. If the problem itself is weakly evidenced, collect enough local evidence to decide whether the change is worth pursuing. If the practice is too vague, return to the active ingredient. If no feasible route closes the gap, reconsider adoption. Readiness is useful because it makes delay purposeful. ## Readiness can differ across parts of the programme A programme does not have to be either completely ready or completely unready. Primary Mathematics may have strong materials and trained tutors. Secondary English may need subject-specific examples. Online tuition may have an unresolved response-mode issue. One branch may have coaching capacity that another lacks. The programme can stage implementation. This is not inconsistency if the staging follows explicit conditions. “Mathematics launches now; English enters a two-week material-development and rehearsal phase; online groups wait until the shared whiteboard access issue is resolved.” That is more credible than declaring programme-wide implementation and accepting hidden exceptions. Reach should be judged against declared eligibility, not an imaginary universal rollout. ## Readiness is temporary A programme can be ready in September and not ready in January. Tutors change. Curriculum materials change. Technology changes. A new branch opens. A practice expands to younger learners. Coaching capacity falls. A policy changes the data that can be collected. Readiness therefore belongs at material transitions. Do not repeat the full launch review constantly. Reopen the relevant parts when the implementation context changes enough to threaten the assumptions on which readiness depended. This connects to the Implementation-Sustainability Gate. Sustainable practice needs a support system that can notice when the conditions that once made implementation viable have changed. ## What tutors should experience when a programme is ready Tutors should be able to answer a small set of questions without guessing. Why are we doing this? What must remain true? When should I use it? What can I adapt? What should I notice in the learner? Where do I go if it does not fit? What evidence are we collecting and why? Who decides whether the practice changes? They should have access to usable materials, not just policy statements. They should have enough rehearsal for the complexity of the move. They should know that early difficulty will be investigated, not automatically treated as non-compliance. This clarity does not remove professional judgement. It makes professional judgement possible inside a shared implementation system. ## What learners and parents should experience Learners should not feel that a new programme practice appeared randomly or that they are test subjects for an unexplained experiment. Parents do not need an implementation manual, but material changes should have a comprehensible educational purpose. If homework becomes shorter and more targeted, explain the reason. If independent first attempts increase, explain what evidence they protect. If a digital tool is introduced, state what the tutor still checks and what the tool is not allowed to decide. If a practice is being piloted, do not overclaim that it is already proven locally. Good readiness often looks like quiet coherence from the outside. The machinery is backstage. The learner experiences a practice that fits the lesson, has a clear purpose and can be adjusted when evidence says it should. ## Failure modes The enthusiasm launch. Leaders like the idea and begin before the problem, practice or support system is sufficiently defined. The calendar launch. A date becomes the main reason to proceed despite unresolved material gaps. The label-only practice. Tutors are told to implement a broad concept such as “formative assessment” without an observable active ingredient. The capacity-blind launch. Time, material, coaching or technical demands are discovered only after tutors begin improvising. The ownerless practice. Nobody has explicit authority to update, adapt or pause the implementation. The training-equals-readiness failure. Completion of a briefing is treated as proof that tutors can execute the practice live. The measurement overload. The programme creates so much launch data collection that monitoring becomes a competing implementation problem. The accessibility-afterthought. Legitimate access and response conditions are considered only after learners are already excluded or evidence becomes hard to interpret. The resistance shortcut. Tutor concerns are labelled attitude before feasibility, coherence or professional-learning gaps are investigated. The pilot-without-question. A small trial occurs, but nobody knows what uncertainty it was meant to reduce. The no-pause failure. Leaders continue expansion because stopping would feel embarrassing even when readiness assumptions have clearly failed. The permanent-preparation failure. Readiness becomes perfectionism and useful implementation never begins. ## Evidence boundaries AERO’s How ready are we? Preparing to start an implementation effort, published and updated 15 September 2026, is a research-informed practice resource for school leaders. It emphasises preparedness, deliberate planning, adequate resourcing, willingness and capacity. It does not provide a validated readiness score for private tutoring. AERO’s Context first: Tailoring implementation to each school setting, also published 15 September 2026, explicitly states that implementation is not one size fits all and should be adjusted to local context and needs. It uses school examples and should be translated cautiously to tutoring. EEF’s A School’s Guide to Implementation, third edition published 24 April 2024, provides a structured but flexible Explore–Prepare–Deliver–Sustain process. It is used here to support the separation between preparation and delivery, not as a mandate for a specific tuition-centre bureaucracy. The seven-condition readiness framework in this article is a professional implementation proposal. Its value should be judged locally: does it reduce predictable rollout failure, make tutor expectations clearer, protect learners from avoidable confusion and produce implementation evidence that can actually guide the next decision? ## The end state A ready programme has not eliminated uncertainty. It has organised uncertainty well enough to learn responsibly. The problem is real enough to justify attention. The practice is clear enough to recognise. The context has been inspected. The main materials and supports exist. Tutors know what they are learning and where to take problems. The programme has a small evidence plan. Ownership is explicit. Someone can adapt, narrow or pause the effort when reality disagrees with the plan. Then implementation can begin. If those conditions are not sufficiently true, more exhortation will not create them. Prepare the missing condition. Test the uncertain part. Narrow the scope. Build the professional-learning route. Resolve the access boundary. Or decide that this is not the right practice now. That is the Implementation-Readiness Gate. A programme should not ask tutors to prove a rollout works before the programme has done enough work to make a fair rollout possible.