Implementation reach, intervention reach, programme reach, student access, participation, coverage, implementation outcomes, educational equity, learning support, intervention dosage, school implementation and evidence-based practice all circle around a deceptively simple question: did the training that was supposed to help actually reach the learner, context and moment that needed it?

A tuition programme can report full enrolment and still have weak reach. Every student may attend, every worksheet may be distributed, every lesson may appear on the calendar, yet the active learning mechanism may never reach the exact performance condition where the weakness occurs. Mira may practise algebra in tuition but not encounter the mixed-method choice that fails in school examinations. Jonas may receive excellent inference feedback during a lesson but never use the evidence-checking routine when he writes alone at home. Nadia may master experimental reasoning in familiar worksheets and lose it when a new apparatus appears under time pressure.

This is why training reach must be defined more carefully than attendance. Current implementation guidance from the Australian Education Research Organisation explicitly pairs fidelity and reach as outcomes worth monitoring. Reach is not merely whether people were present. It concerns whether implementation touches the intended participants and whether the intended practice is actually experienced. In a family-and-tuition system, the useful question is even more precise: does the active ingredient reach the learner in the places, task types and conditions where it must eventually work?

Training reach is the extent to which the intended training mechanism reaches the relevant learner, capability, setting, task condition and performance moment often enough to create, verify and sustain the intended change.

Reach is about coverage, but not only population coverage.

It is also capability coverage.

Context coverage.

Condition coverage.

Time coverage.

Transfer coverage.

And, eventually, ownership coverage: whether the learner can carry the mechanism when the tutor is no longer present.

---

Quick Read: Reach Is More Than Showing Up

Identify Who and What Need the Intervention → Define the Active Ingredient → Map Where It Must Operate → Check Who Actually Receives It → Check Which Conditions It Reaches → Find the Gaps → Adapt Delivery → Verify Reach in Authentic Performance

A useful reach check asks:

• Did the learner receive the active ingredient?
• Did all relevant learners receive it, or only the easiest-to-support learners?
• Did it reach the weak capability rather than a neighbouring easier skill?
• Did it reach home, school and tuition where necessary?
• Did it reach independent performance, or only supported practice?
• Did it reach unfamiliar tasks, or only familiar examples?
• Did it reach high-pressure conditions when those conditions matter?
• Did it reach often enough and for long enough to matter?
• Did the learner internalise it, or does the mechanism disappear outside adult supervision?

The Current Apex Implementation Frame

AERO’s 15 September 2026 practice guide Staying on track: Monitoring implementation outcomes explicitly groups fidelity and reach among the implementation outcomes school teams should monitor. That pairing is valuable because an intervention can be delivered well to the learners who receive it and still fail at system level if important learners or settings are not reached.

AERO’s companion guide Context first: Tailoring implementation to each school setting emphasises that implementation should be adapted to local contexts rather than applied as one fixed template. That matters to reach because a delivery form that works in one context may never meaningfully reach learners in another.

EEF’s 2026 guide on adaptations and additional support similarly asks what adaptations or additional support some pupils may need, while cautioning that adaptations should be judged by whether they help rather than hinder learning. The reach lesson is straightforward: universal access and targeted support are both implementation questions.

At the larger systems level, a 2025 open-access paper in Global Implementation Research and Applications describes work on developing implementation and scaling capacity across education systems. Scaling increases the challenge of maintaining implementation support across diverse settings. Reach is therefore not a trivial afterthought. It is a core property of whether an intervention exists beyond its original pilot conditions.

Reach Owner Boundary: What This Article Owns

This article owns one specific job:

Does the intended training mechanism actually reach the learner, capability, setting and performance conditions where it is supposed to operate?

It does not re-own several adjacent mechanisms.

• Training Fidelity asks whether the mechanism was delivered as intended when it was delivered.
• Training Feasibility asks whether the plan can operate under real constraints.
• Training Acceptability asks whether learners, families and tutors will actually use the plan.
• Training Sustainability asks whether the system survives over time.
• Training Handoffs owns the movement of capability and responsibility among teacher, tutor, parent and learner.
• Distribution Shift owns the broader problem of practice and performance coming from different distributions.

Reach is the coverage layer between them.

Reach Level 1: Does the Learner Receive the Intervention at All?

The most basic reach question is literal.

Was the learner present?

Did the task arrive?

Did the session occur?

Did the learner open the material?

Did the intervention reach the person for whom it was intended?

This is necessary.

It is not sufficient.

Attendance Is a Reach Proxy, Not Reach Itself

A student can attend every tuition lesson and never experience the critical training operation.

Mira attends a 1.5-hour lesson.

The target is method selection under mixed conditions.

The session contains only blocked algebra practice.

Attendance reach is high.

Mechanism reach is low.

Worksheet Distribution Is a Reach Proxy

Every learner receives the worksheet.

Some complete it.

Some copy.

Some use model answers too early.

Some skip the section that contains the intended active ingredient.

Distribution tells us where paper travelled.

It does not tell us where learning operations travelled.

Platform Access Is a Reach Proxy

A digital platform can report that a learner logged in.

That does not prove:

• the target task was opened;
• the learner attempted before help;
• the learner used feedback;
• the learner reached the difficult item;
• the learner completed a fresh retest.

Digital traces can improve reach monitoring, but only if the trace corresponds to the mechanism.

Reach Level 2: Does the Active Ingredient Reach the Learner?

Training plans contain active ingredients.

Contrast.

Retrieval.

Fresh retest.

Evidence selection.

Method classification.

Independent initiation.

Reach improves when we ask whether these operations reached the learner rather than whether generic study occurred.

Active-Ingredient Reach Example: Mathematics

Mira’s target is mixed method selection.

Active ingredient:

See competing cases → classify before solving → state discriminating cue → solve → fresh mixed retest.

If Mira receives four pages of quadratics but every page labels the method, quantity reach is high and mechanism reach is low.

Active-Ingredient Reach Example: English

Jonas’s target is evidence-bounded inference.

Active ingredient:

select evidence → propose claim → compare claim strength with evidence → revise → fresh passage.

If Jonas receives ten inference questions but the tutor supplies the evidence line every time, answer volume reaches him while independent evidence selection does not.

Active-Ingredient Reach Example: Science

Nadia’s target is evidence-to-mechanism reasoning.

Active ingredient:

identify evidence → state pattern → connect mechanism → bound conclusion.

If Nadia memorises model answers, Science language reaches her.

The intended reasoning chain may not.

Reach Level 3: Does the Training Reach the Weak Capability?

A training system can reach the learner but miss the bottleneck.

Mira receives more algebra when the actual weakness is representation translation.

Jonas receives grammar practice when the actual weakness is idea development.

Nadia receives concept revision when the actual weakness is evidence integration.

This is capability misreach.

Capability Misreach Often Looks Productive

The child works.

Pages fill.

Adults see effort.

The weak link remains untouched.

Because nearby tasks are easier to assign, systems naturally drift toward them.

Reach requires diagnosis before distribution.

Reach Level 4: Does the Training Reach the Right Difficulty Band?

A task can reach the right capability and still sit outside the learner’s useful range.

Too easy creates ceiling effects.

Too hard creates floor effects.

Reach therefore includes difficulty access.

The learner needs training opportunities near enough to current capability that the target operation can actually occur.

Difficulty Reach and Ceiling Effects

Mira has mastered straightforward factorisation.

More easy factorisation reaches a capability she already owns.

It does not reach the boundary where learning is still needed.

Training Ceiling Effects describes this measurement problem.

Reach translates it into implementation: training should extend into the learner’s active growth range.

Difficulty Reach and Floor Effects

A learner receives only maximum-difficulty exam tasks.

They fail before the target reasoning operation can even begin.

The intervention reaches the learner physically but not cognitively.

Use decomposition, pre-training or cueing to move the task into an operable range.

Reach Level 5: Does the Training Reach Different Representations?

Knowledge can remain representation-bound.

An algebraic relationship works in equations and disappears in graphs.

An inference rule works in narrative and disappears in speeches.

An experimental reasoning routine works in plant diagrams and disappears in electricity setups.

Reach therefore includes representation coverage.

Representation Reach Without Redundancy

The aim is not to show every possible representation.

Select enough forms to reveal whether the learner carries the invariant relationship.

This connects with the existing HPL owner on representation switching without duplicating it.

Reach Level 6: Does the Training Reach Independent Performance?

A capability can live entirely inside supported tuition.

The tutor asks the first question.

The tutor names the category.

The tutor notices the error.

The learner performs well.

At home or school, the system disappears.

This is support-bound reach.

Independent Reach Is the Real Handoff Test

Eventually the mechanism must cross from external support to learner control.

The question is not merely:

Can Mira solve this with the tutor?

It is:

Does the method-selection routine reach Mira’s own thinking when nobody prompts it?

Reach Level 7: Does the Training Reach Home?

Home is not a second tuition centre.

Reach at home may be light:

• one retrieval return;
• one fresh retest;
• one self-check routine;
• one planned writing paragraph.

The question is whether enough of the mechanism reaches home to support continuity and independence.

Home Reach Failure: Parent Rescue

Mira cannot begin.

Parent immediately names the method.

Homework is completed.

Independent method-selection reach remains low.

The family may need a shared cue rule:

Ask what the question is asking; do not name the method.

Reach Level 8: Does the Training Reach School Performance?

The tuition system ultimately has to travel into school tasks.

If a checking routine exists only in tuition, reach is incomplete.

If a paragraph-planning method disappears in timed school writing, reach is incomplete.

If a Science evidence routine survives only on tutor-selected diagrams, reach is incomplete.

School Reach Should Be Tested, Not Assumed

Look at school papers.

Ask whether the trained process appears naturally.

Do not assume transfer simply because the learner can explain the routine in tuition.

Reach Level 9: Does the Training Reach Examination Conditions?

Authentic performance conditions matter.

Time.

Mixed topics.

No prompts.

Unfamiliar item order.

Emotional pressure.

A skill that reaches only calm practice is not yet examination-ready.

Examination Reach Should Be Added Progressively

Do not introduce full pressure too early.

Sequence:

• accurate supported practice;
• independent untimed practice;
• fresh mixed practice;
• relevant timing;
• whole-paper simulation.

Reach broadens as readiness grows.

Reach Level 10: Does the Training Reach After Delay?

Immediate availability can disappear.

A mechanism that works only while the explanation is warm has short temporal reach.

Use delayed returns.

Training Retests owns the verification loop.

Reach asks whether the loop extends across time.

Temporal Reach

Some training is intended to last minutes.

Some weeks.

Some years.

The required temporal reach should match the future demand.

A formula needed throughout Secondary Mathematics deserves longer temporal reach than a one-off class-specific convention.

Reach Level 11: Does the Training Reach Under Fatigue?

Performance changes with fatigue.

That does not mean every skill should be trained in exhaustion.

But if the real task often occurs late in a long examination, robustness may eventually need testing under realistic load.

Fatigue Reach Is a Later-Stage Test

First build the skill in normal conditions.

Then test graceful degradation.

The existing HPL article on Graceful Degradation owns the broader robustness mechanism.

Reach Level 12: Does the Training Reach When the Surface Changes?

A learner may succeed on familiar examples and fail when the context changes.

This is surface-limited reach.

Use case families, perturbation and transfer checks.

Reach Level 13: Does the Training Reach the Learners Most Likely to Be Missed?

At group level, implementation can systematically favour learners who are easiest to teach.

Confident students volunteer.

Fast workers reach extension tasks.

Quiet students remain hidden.

Students who are absent during key instruction miss the active ingredient.

Reach should identify who is systematically under-served.

Reach and Educational Equity

Equity does not mean identical training.

It means relevant learners can access the learning opportunities they need.

A universal lesson may need targeted additional support for some learners.

EEF’s 2026 inclusive teaching guidance makes this explicit: universal approaches can be complemented by adaptations and additional support when needed.

Reach and Absence

Absence creates reach gaps.

Do not solve every absence with full backlog.

Ask:

• Which active ingredient was missed?
• Which dependency matters next?
• What minimum re-entry task restores reach?

This connects to Training Re-entry.

Reach and Quiet Learners

Quiet students can appear low-maintenance.

They may be under-reached if help is distributed to whoever asks loudly.

Use deliberate sampling.

Check cold starts.

Inspect working.

Reach and High Performers

High performers can be under-reached too.

If every task remains below their useful challenge range, they receive instruction but little growth opportunity.

Ceiling-aware reach matters.

Reach and Learners Who Need More Time

If every lesson moves on before the learner finishes the target operation, nominal coverage can hide low reach.

Give enough processing time for the mechanism to occur.

Reach and Language Background

Instructions can block reach.

If vocabulary is incidental to the target, clarify it.

If vocabulary is the target, preserve it.

Adapt delivery so irrelevant language difficulty does not prevent the intended capability from being exercised.

Reach and Accessibility

Font size, colour contrast, audio access, physical setup and response format can determine whether training reaches a learner.

Accessibility is not separate from implementation.

Reach and Neurodiversity

Known learner needs may require predictable structure, response-format adaptations or sensory adjustments.

Do not infer diagnoses from performance.

But do ensure delivery conditions do not unnecessarily block access to the target learning operation.

Reach Level 14: Does the Training Reach the Correct Moment?

Timing matters.

Feedback too early can remove a decision.

Feedback too late can become disconnected from the attempt.

Pre-training after the complex task is not pre-training.

A retest before feedback does not verify repair.

Reach has a temporal position inside the learning sequence.

Moment Reach Example: Cueing

A cue should enter after productive struggle begins but before unproductive collapse.

If it enters immediately, the learner never makes the decision.

If it enters after ten minutes of random search, time is wasted.

Moment Reach Example: Feedback

Feedback should reach the learner while the relevant mental model is still recoverable and before the error becomes repeated habit.

But the learner may need an initial independent attempt first.

Moment Reach Example: Retest

The fresh retest should occur after repair, not before.

A delayed retest should occur after enough time has passed to require reconstruction.

Reach Level 15: Does the Training Reach the Authentic Decision?

Many exercises practise operations but remove decisions.

Chapter headings name methods.

Question groups reveal categories.

Model paragraphs reveal structures.

Teacher prompts reveal what matters.

Reach requires the training to eventually include the decision the learner must make independently.

The Decision-Reach Test

If we removed the teacher, chapter title and example label, would the learner still know what decision to make?

If not, decision reach remains incomplete.

Reach Level 16: Does the Training Reach the Learner’s Self-Regulation?

Eventually the learner must decide:

• when to start;
• what to practise;
• when to ask for help;
• when to check;
• when to stop.

Training reach becomes self-regulatory when the learner carries those control rules outside the training environment.

Reach and Prompt Migration

The site’s Prompt Migration article gives a concrete reach path.

Tutor prompt → shared prompt → learner self-prompt.

The cue’s reach expands from one classroom to the learner’s own future performance.

Reach and the Cold Start Test

The Cold Start Test asks whether the learner can begin without tutor activation.

It is a reach test for initiation control.

Reach and Training Handoffs

Training Handoffs asks how capability moves between people and settings.

Reach asks whether the mechanism actually arrives after the handoff.

A tutor can explain the home routine perfectly to a parent.

If the learner never uses it at home, the handoff occurred administratively and reach remained incomplete.

Reach and Training Fidelity

Reach and fidelity answer different questions.

Fidelity: did the learner who received the intervention receive the intended mechanism?

Reach: did the mechanism reach all relevant learners, contexts and conditions where it was needed?

High fidelity + low reach:

a beautiful intervention delivered to too few conditions.

High reach + low fidelity:

everyone receives something, but the mechanism is diluted.

Strong implementation needs both.

Reach and Feasibility

Training Feasibility determines whether reach can be sustained.

A plan that requires resources unavailable at home will have limited home reach.

A plan that assumes parental supervision every evening may have low reach when parents work late.

Reach and Acceptability

If a learner rejects the format, reach shrinks.

The active ingredient may be theoretically available but practically unused.

Adapt form without breaking function.

Reach and Sustainability

Initial reach can be high during launch.

Later reach can shrink as reminders, materials and adult attention fade.

Training Sustainability protects reach across time.

Reach and Adaptation

Different learners and settings may require different delivery forms to achieve equivalent mechanism reach.

The companion article Training Adaptation owns the question of how to change delivery without breaking the active ingredient.

Reach and Scale

As a successful intervention spreads across learners, classes or sites, reach usually becomes harder to guarantee.

Coverage expands.

Support intensity may fall.

Context variation rises.

The companion article Training Scale owns that expansion problem.

The Reach Denominator Problem

To measure reach, we need to know who or what should have been reached.

If five students receive a support programme, is reach 100%?

Only if five were the complete relevant group.

If ten needed it, reach may be closer to half.

The denominator matters.

Individual Training Has Denominators Too

Mira has six recurring task conditions where method selection matters.

Practice reaches two.

It would be misleading to say “method selection has been trained” without noting the condition coverage.

The Reach Map

For an active target, map five dimensions:

• Who: which learner or subgroup?
• What: which capability?
• Where: which setting?
• When: which point in the learning/performance sequence?
• Under what conditions: support, time, representation, difficulty, novelty?

Any empty cell may be a reach gap.

A Reach Matrix for Mira

Target: quadratic method selection.

• tuition, untimed, mixed, supported — reached;
• tuition, untimed, mixed, independent — partly reached;
• home, mixed, independent — weak reach;
• school homework, topic-labelled — high practice, low selection reach;
• exam, timed, mixed — currently low reach;
• fresh unfamiliar representation — not yet reached.

The map tells the tutor what to expand next.

A Reach Matrix for Jonas

Target: evidence-bounded inference.

• narrative passages — reached;
• articles — partly reached;
• speeches — not yet reached;
• supported evidence selection — reached;
• independent evidence selection — improving;
• timed exam conditions — not yet tested;
• home writing transfer — uncertain.

A Reach Matrix for Nadia

Target: evidence-to-mechanism reasoning.

• familiar experiments — reached;
• unfamiliar apparatus — partly reached;
• graph evidence — reached;
• table evidence — reached;
• timed explanations — weak reach;
• new Science topic — not yet tested.

Reach Gaps Are Not Automatically Failures

Early training should have narrow reach.

A novice cannot handle every context at once.

The issue is whether reach broadens appropriately over time.

Start narrow.

Stabilise.

Then expand.

The Reach Expansion Sequence

A useful general sequence:

• clean example;
• several examples;
• contrast;
• fresh parallel item;
• changed representation;
• mixed context;
• delay;
• reduced support;
• authentic time pressure;
• whole-task performance.

Not every skill needs every stage.

The sequence illustrates widening reach.

Reach and Training Case Families

Training Case Families provides a practical way to extend reach systematically.

A case family can include:

• prototype;
• near-miss;
• boundary case;
• representation change;
• distractor-rich case;
• transfer case.

The tutor expands reach without abandoning the underlying concept.

Reach and Perturbation

Training Perturbation changes one feature at a time.

That is ideal for identifying exactly which context change breaks reach.

Reach and Invariants

Training Invariants helps learners identify what remains stable as reach broadens.

Reach and Nonexamples

Nonexamples extend reach to the concept boundary.

The learner learns not only where the rule applies but where it stops.

Reach and Ambiguity

Ambiguous cases extend reach into conditions where multiple interpretations remain plausible.

This should come after clearer cases are stable.

Reach and Distractors

Plausible distractors extend reach into competitive decision environments.

Random wrong answers do not.

Reach and Interference

Reach can fail when a neighbouring rule wins.

Training Interference owns that mechanism.

Reach planning should include cases where competing rules are present.

Reach and Re-entry

After a break, previous reach may contract.

Do not assume all contexts remain stable.

Sample.

Then rebuild reach where needed.

Reach and Transition

A capability that reached Primary contexts may not automatically reach Secondary demands.

More abstraction.

More independence.

Different representations.

Longer papers.

Reach must be re-tested when performance conditions change.

Reach and Distribution Shift

The site’s HPL article on Distribution Shift explains what happens when performance stops coming from the same world as practice.

Training Reach offers an implementation response:

expand the training distribution deliberately enough that important future worlds are not entirely unseen.

Reach and External Validity

The site’s Scientific External Validity owner asks whether a result travels.

Training Reach stays practical and individual:

did this learner actually receive enough varied, authentic opportunities for the capability to operate where it matters?

Reach and Transportability

Transportability is a scientific evidence question.

Reach is an implementation-delivery question.

Do not merge them.

Reach and Equity: The Easy-to-Reach Learner Bias

Systems naturally serve learners who:

• attend consistently;
• ask questions;
• finish quickly;
• have strong parent support;
• have devices and quiet spaces;
• understand instructions quickly.

Reach monitoring should specifically inspect learners who do not fit those conditions.

Reach and the Family Attention System

The site’s Family Attention System reminds us that family attention is finite.

A child may be under-reached at home because attention is consumed by siblings, work or caregiving.

Design home training that does not assume unlimited supervision.

Reach and the Sibling System

Shared devices, quiet spaces and parent time can create reach differences between siblings.

Make resource allocation explicit rather than accidental.

Reach and Family Coordination

Transport and schedules can determine whether a learner reaches tuition at all.

But deeper reach also asks whether the training job survives the handoff from tuition into home and school.

Reach and Family Motivation

Low family motivation can reduce use, but do not diagnose motivation before checking feasibility and acceptability.

A system can be under-reached because it is too burdensome, not because the family does not care.

Reach and Family Standards

Household standards can protect a minimum study routine.

But standards should define the core behaviour clearly enough that the intended mechanism—not generic “study time”—reaches the learner.

The Reach Funnel

At each stage, potential reach can shrink.

Eligible Learners → Offered Support → Attend → Begin Task → Perform Active Ingredient → Complete Dose → Transfer to Fresh Task → Use Independently → Use in Authentic Performance

Counting only the first or second stage overstates real reach.

Reach Funnel Example: Tuition Homework

Ten learners receive homework.

Eight complete it.

Six complete before seeing solutions.

Five perform the target retrieval operation.

Four repair errors.

Three do a fresh retest.

If the mechanism required fresh retest, the meaningful reach may be closer to three than ten.

Reach Funnel Example: One Learner

Mira has ten planned mixed decisions.

She attempts eight.

Two are effectively labelled by the worksheet.

One receives a full tutor cue before selection.

Five are genuine independent method decisions.

The meaningful dose is five.

Reach Should Not Become Surveillance

We do not need to track every micro-step.

Use the smallest reach indicators that answer the implementation question.

Reach Indicator 1: Exposure

Did the learner encounter the training opportunity?

Reach Indicator 2: Participation

Did the learner actually perform the target operation?

Reach Indicator 3: Dose

Did enough meaningful opportunities occur?

Reach Indicator 4: Context Coverage

Which relevant conditions have been sampled?

Reach Indicator 5: Independence

Does the mechanism appear without external activation?

Reach Indicator 6: Authentic Use

Does it appear in real school or examination performance?

Reach Indicators Can Be Qualitative

A note can be enough:

Evidence-checking routine appears reliably in tuition, inconsistently in home comprehension and not yet under timed school conditions.

That is an actionable reach map.

Do Not Invent False Reach Percentages

“73% reach” may look scientific and mean very little if the denominator is unclear.

Use categories when data are sparse:

• not reached;
• emerging reach;
• reached under support;
• independent reach;
• authentic reach.

Reach and Evidence Thresholds

How much reach is enough before progressing?

That depends on risk.

A low-stakes skill may progress after modest coverage.

A high-stakes examination routine may need broader condition reach.

The Evidence Threshold owns the decision rule.

Reach and Exit Criteria

A skill should not leave active training because it worked in one narrow condition.

Exit criteria should specify the reach that matters:

• fresh tasks;
• delay;
• relevant variation;
• independence;
• authentic conditions.

Reach Failure Mode: Attendance Equals Reach

Student was present.

Therefore intervention reached them.

Repair:

check the active operation.

Reach Failure Mode: Homework Completion Equals Reach

Worksheet finished.

Therefore practice happened.

Repair:

check whether the target decision, retrieval or repair actually occurred.

Reach Failure Mode: One Context Equals General Reach

Learner succeeds in tuition.

Adults assume school transfer.

Repair:

sample authentic context.

Reach Failure Mode: The Easy Learners Get the Best Intervention

Fast students reach the extension or fresh retest.

Struggling students spend the whole lesson finishing basic work.

Repair:

protect active-ingredient time for the learners who need it most.

Reach Failure Mode: Parent Support Determines Access

Only learners with active parental supervision complete the home component.

Repair:

redesign the home loop for learner independence.

Reach Failure Mode: Digital-Only Delivery

Device or login friction blocks access.

Repair:

provide low-friction fallback.

Reach Failure Mode: Tutor-Only Mechanism

Routine disappears outside tuition.

Repair:

prompt migration and independent retest.

Reach Failure Mode: Practice Never Reaches Time Pressure

Accuracy is strong untimed.

Exam execution collapses.

Repair:

add timing after accuracy and independence stabilise.

Reach Failure Mode: Practice Never Reaches Transfer

Repeated familiar examples create confidence.

Novel context breaks performance.

Repair:

expand case family.

Reach Failure Mode: Practice Reaches Too Far Too Soon

Adult jumps immediately to far-transfer tasks.

Learner cannot detect the invariant.

Repair:

stabilise nearer cases first.

Reach Failure Mode: Every Context Is Trained Directly

Adults try to prepare the learner for every possible question surface.

This is impossible.

Repair:

train invariants and representative variation.

Reach Failure Mode: No Denominator

Adults report “good coverage” without defining what conditions mattered.

Repair:

map relevant contexts and active ingredients.

Reach Failure Mode: No Low-Capacity Path

Busy weeks eliminate the intervention entirely.

Repair:

create minimum viable reach.

Minimum Viable Reach

When capacity is low, what is the smallest contact with the active ingredient that prevents total disappearance?

Examples:

• three method classifications;
• one evidence-to-claim comparison;
• one Science explanation chain;
• five retrieval prompts;
• one fresh retest.

This is not ideal dose.

It preserves continuity.

Reach and Graceful Degradation

When the week worsens, reduce quantity before removing the core mechanism.

This extends reach through difficult periods.

Reach and Sustainability

Sustainable systems maintain enough reach after novelty fades.

If the home component disappears every term, it never became sustainable reach.

Reach and Replanning

If reach repeatedly fails in one setting, use The Replanning Trigger.

Maybe the setting, timing or delivery form must change.

The Reach Review Cycle

Every few weeks, ask:

• What active targets exist?
• Where are they currently reached?
• Where do they still disappear?
• Which new context matters next?
• Which context no longer needs dedicated practice?
• What support should migrate?

Reach Should Expand and Contract

As a skill stabilises, dedicated reach can contract.

The capability should continue through authentic work while the intervention shrinks.

Sustainability does not mean permanent maximum reach of a special programme.

Reach and De-implementation

When a routine no longer needs dedicated reach, remove it from active training.

The companion De-implementation article owns how low-value routines should leave the system.

Reach and Scale

If a mechanism works for Mira, can it reach Jonas and Nadia without becoming generic?

Scaling should not mean giving identical materials to everyone.

Scale should preserve access to the relevant active ingredient while adapting delivery.

The Reach–Fidelity Matrix

• High reach + high fidelity: strong implementation coverage.
• High reach + low fidelity: diluted implementation.
• Low reach + high fidelity: strong local intervention, narrow coverage.
• Low reach + low fidelity: implementation breakdown.

The Reach–Feasibility Matrix

• High feasibility + low reach: likely design, targeting or adoption problem.
• Low feasibility + low reach: redesign constraints.
• High feasibility + high reach: good operational coverage.
• Low feasibility + temporarily high reach: watch for burnout.

The Reach–Acceptability Matrix

• High acceptability + low reach: people like the plan but cannot access it fully.
• Low acceptability + low reach: resistance may be blocking use.
• High reach + low acceptability: compliance may be fragile.
• High reach + high acceptability: strong uptake, still requiring outcome evidence.

The Reach–Outcome Matrix

• High reach + improvement: mechanism plausibly contributes.
• High reach + no improvement: review fidelity, validity and plan quality.
• Low reach + improvement: other causes or partial mechanism may explain change.
• Low reach + no improvement: implementation too weak for clean method judgement.

Reach and Counterfactual Thinking

If a capability improves in a context the intervention never reached, the intervention cannot be the only explanation.

Reach therefore constrains causal stories.

Reach and the Flight Recorder

The Flight Recorder can log not only performance history but where the mechanism was actually used.

Example:

Inference calibration reached tuition and home tasks; not yet observed in timed school paper.

Reach and the Learning Dispatcher

The Learning Dispatcher sends a learner to the right intervention.

Reach asks whether the intervention then actually enters the learner’s practice environment.

Reach and the Translation Layer

A method may fail to reach because language differs between school, tuition and home.

Translate terminology while preserving concept.

Reach and the Evidence Threshold

Large implementation expansions require evidence.

Do not add home, school and weekend practice simply because one context remains weak.

Identify which reach gap materially limits the outcome.

Reach and the Counterfactual Check

If improvement appears only in the intervention context, broader reach may still be incomplete.

Counterfactual reasoning should include setting coverage.

Reach and Family-Life Education

eduKatePunggol’s local owner layer matters here.

Training is not delivered into a vacuum.

The mechanism must cross:

• school timetable;
• transport;
• home routines;
• tuition schedule;
• parent support;
• sibling competition for resources;
• learner energy;
• exam cycles.

Family life determines which reach routes are viable.

Reach Example: Wednesday Evening in Punggol

Mira returns home after school and CCA.

She has thirty usable minutes.

The ideal plan contains twenty mixed Mathematics questions.

Low-capacity reach design uses:

• three method classifications;
• one full solve;
• one fresh check.

The mechanism reaches the day even when full dose cannot.

Reach Example: Parent Works Late

Jonas’s home inference routine originally requires parental checking.

Parent availability changes.

Reach falls.

Redesign:

• answer key with evidence anchors;
• self-check rubric;
• flag unresolved items for tuition.

Mechanism reach becomes less parent-dependent.

Reach Example: Nadia’s Busy Exam Week

Full Science papers are infeasible.

Keep one evidence-to-mechanism chain each alternate day.

The active ingredient remains in contact with the learner.

Reach Example: Evan’s Peer Study

Evan and Mira study together.

Peer comparison can extend reach socially.

But if Evan reveals the answer before Mira commits, decision reach shrinks.

Use a sequence:

attempt alone → compare → explain difference → fresh individual check.

Reach and AI-Assisted Study

AI can extend reach by making feedback, examples and explanations available outside tuition.

But AI can also reduce reach of learner-generated operations.

If Mira asks the AI to choose the method, method-selection practice disappears.

If Jonas asks for the inference, evidence selection disappears.

If Nadia asks for the full explanation, causal generation disappears.

AI Reach Rule: Extend Support, Not Substitute the Target Operation

Use AI to reach contexts the tutor cannot:

• generate fresh examples;
• provide delayed feedback;
• offer vocabulary clarification;
• create parallel forms;
• schedule returns.

Protect unaided attempts where generation, retrieval or decision-making is the job.

AI Reach Rule: Verify Authentic Transfer

A learner may perform strongly with AI support and weakly without it.

Reach to exam conditions requires tool-free verification when the examination is tool-free.

Reach and Privacy

Extending reach digitally should not require unnecessary data collection.

Use the minimum data needed to support learning.

Reach and Resource Equity

Digital interventions can widen reach for some and narrow it for learners with limited devices, connectivity or quiet space.

Provide alternative delivery paths.

Reach and Tutor Capacity

The tutor cannot deliver high-intensity individual support to unlimited learners.

Reach depends on capacity.

Three-student small-group design protects a particular reach model:

• shared explanation;
• individual observation;
• individual branching;
• fresh retesting;
• peer comparison.

Reach and Scaling Class Size

Add more students and the same lesson may no longer reach individual weak links.

That does not mean larger classes cannot work.

It means the implementation architecture must change.

Reach and Teacher Attention

Attention itself has reach.

Who gets feedback?

Who gets cold-start testing?

Who gets fresh retests?

Who is invisible because they are quiet?

Monitor attention distribution.

Reach and Response Opportunity

In a group lesson, every learner should have enough opportunities to perform the target operation.

Watching another student perform is not identical to performing.

Reach and Peer Observation

Observation can still contribute if learners predict before the peer answers.

This turns shared discussion into distributed cognitive participation.

Reach and Worked Examples

Worked examples can reach several learners simultaneously.

But transition to independent completion must reach each learner individually.

The eduKateSengkang Worked Example State remains the canonical progression owner.

Reach and Feedback Distribution

One learner’s error can become a group learning opportunity if anonymised and relevant.

But individual feedback must still reach the learner whose specific weak link differs.

Reach and Differentiation

Differentiation is not giving every learner a different worksheet.

It is ensuring the relevant training operation reaches each learner.

Reach and Shared Curriculum

Three learners can share the same topic while receiving different reach extensions:

• Mira — method selection;
• Jonas — execution stability;
• Nadia — transfer.

Same curriculum surface.

Different active ingredient.

Reach and Universal Approaches

Some practices should reach everyone:

• clear explanations;
• opportunities to practise;
• feedback;
• retrieval;
• appropriate challenge.

Additional targeted support extends reach where universal provision is insufficient.

Reach and Targeted Support

Targeted support should have an entry rule and an exit rule.

Otherwise reach becomes permanent additional workload.

Reach and Stigma

Targeted support can reduce acceptability if learners feel labelled.

Normalise branching and private diagnostics.

Reach and Success

When the learner improves, reach should change.

Do not keep maximum support everywhere.

Move from supported reach to independent reach.

Reach and Maintenance

After stability, reach can become lighter and more distributed:

• cumulative mixed items;
• monthly anchor task;
• natural use in schoolwork.

Reach and De-implementation

Remove dedicated interventions when ordinary work now carries the capability adequately.

Reach and New Weak Links

Freed capacity can reach the next bottleneck.

This is why exit is part of reach management.

The Reach Portfolio

A learner has limited training capacity.

Think of reach as a portfolio:

• which active targets currently need direct contact?
• which need maintenance only?
• which authentic conditions need testing?
• which contexts can be left to ordinary curriculum?

Do Not Maximise Reach Everywhere

Maximum reach is not the goal.

Relevant reach is.

Overextending every skill into every context wastes capacity.

The Reach Stop Rule

Stop expanding dedicated reach when:

• the skill survives representative variation;
• independent performance is stable;
• authentic conditions are covered;
• maintenance can take over.

The Reach Renewal Rule

Renew dedicated reach when:

• context changes;
• performance drifts;
• a transition increases demand;
• new interference appears.

Reach Failure Mode: Reaching the Proxy, Not the Capability

Worksheet completion rises.

Capability does not.

This connects to the HPL Proxy Failure owner.

Reach Failure Mode: Reaching Practice, Not Performance

Routine works in training and disappears in exam.

Expand to authentic conditions.

Reach Failure Mode: Reaching One Surface

Skill remains tied to one representation.

Vary surface deliberately.

Reach Failure Mode: Reaching Only When Prompted

Mechanism exists only after adult cue.

Migrate cue.

Reach Failure Mode: Reaching Too Many Low-Value Contexts

Training expands everywhere and becomes burdensome.

Prioritise representative conditions.

Reach Failure Mode: Missing the High-Stakes Context

Training covers homework but not examination timing.

Add authentic simulation when ready.

Reach Failure Mode: Missing the Low-Stakes Context

Skill appears only under exam pressure because routine practice is absent.

Build ordinary-week maintenance.

Reach Failure Mode: Reach Depends on One Adult

Parent or tutor absence collapses use.

Build learner ownership.

Reach Failure Mode: Reach Depends on One Tool

Platform unavailable, training stops.

Teach portable routines.

Reach Failure Mode: Reach Is Assumed From Marks

Marks improve.

Adults assume the intervention reached the intended process.

Check fidelity and process evidence.

Reach Failure Mode: Reach Is Assumed From Effort

Learner spends hours.

Target operation remains rare.

Track meaningful attempts.

Reach Failure Mode: The Reach Dashboard Becomes the Goal

System chases attendance, login or completion percentages.

Repair:

connect reach indicators to active ingredients and outcomes.

The Parent Reach Audit

• What skill are we actually trying to reach?
• Does the home task contain that skill?
• Is the child doing it independently?
• Does parent help remove the target decision?
• Does the routine appear at school?
• Does it appear under time pressure when needed?
• Has the skill been tried on fresh examples?
• Has it been tried after delay?
• Which setting still has a gap?
• Is the gap important enough to justify more work?

The Tutor Reach Audit

• Who should receive this intervention?
• Which learners are easiest to reach?
• Who is being missed?
• What is the active ingredient?
• How many meaningful opportunities occurred?
• Which representations were covered?
• Which difficulty bands were covered?
• Which contexts were covered?
• What support level was used?
• Has independent reach been tested?
• Has delay been tested?
• Has authentic school performance been observed?
• What reach expansion is next?
• What dedicated reach can now stop?

The Learner Reach Audit

• Can I use this skill outside tuition?
• Can I use it when the question looks different?
• Can I use it when nobody tells me the method?
• Can I use it after a week?
• Can I use it when I am timed?
• Can I start without a reminder?
• Can I check myself?
• Which situation still breaks the skill?

The Reach Field Manual: Eighty Questions

• Who is the intended learner?
• What capability must be reached?
• What active ingredient carries the mechanism?
• Was it offered?
• Was it accessed?
• Was it attempted?
• Was it completed?
• Was it completed independently?
• Was help required?
• Did the help remove the decision?
• Was the task fresh?
• Was the task familiar?
• Was the method labelled?
• Was the representation familiar?
• Was the difficulty appropriate?
• Was the learner too far below the task?
• Was the learner above the task?
• Did the mechanism reach home?
• Did it reach school?
• Did it reach tuition?
• Did it reach homework?
• Did it reach examination simulation?
• Did it reach actual examination performance?
• Did it reach under time pressure?
• Did it reach without time pressure?
• Did it reach after delay?
• Did it reach with different wording?
• Did it reach with different numbers?
• Did it reach with different context?
• Did it reach a different representation?
• Did it reach a different subject context where relevant?
• Did it reach after a break?
• Did it reach when the tutor was absent?
• Did it reach when the parent was absent?
• Did it reach when AI was unavailable?
• Did it reach when the chapter heading disappeared?
• Did it reach when competing methods were present?
• Did it reach when distractors were plausible?
• Did it reach when the learner was uncertain?
• Did it reach after an error?
• Did it reach during recovery?
• Did it reach during a busy school week?
• Did it reach during holidays?
• Did it reach during exam season?
• Did it reach after illness?
• Did it reach after a school transition?
• Which learners are absent from the intervention?
• Which learners receive less tutor attention?
• Which learners finish too slowly to reach the key task?
• Which learners finish too quickly and never reach useful challenge?
• Which learners lack device access?
• Which learners lack quiet space?
• Which learners need additional language clarification?
• Which learners need alternate response formats?
• Does the denominator include everyone who needs support?
• Are we counting attendance instead of mechanism exposure?
• Are we counting worksheets instead of cognitive operations?
• Are we counting logins instead of participation?
• Are we counting time instead of meaningful attempts?
• What context is still unreached?
• What condition is still unreached?
• What is the next representative reach expansion?
• What expansion would be redundant?
• What expansion would be too hard too soon?
• What active reach can move to maintenance?
• What dedicated intervention can stop?
• What reach gap is actually costing marks?
• What reach gap is merely theoretically interesting?
• What is the minimum viable reach during busy weeks?
• What is the reach plan during high-capacity weeks?
• What support should migrate to the learner?
• What evidence shows authentic reach?
• What evidence would falsify our reach assumption?
• What does school work show?
• What does home work show?
• What does tuition show?
• Where do those windows disagree?
• What should we adapt to improve reach?
• What must stay fixed to preserve fidelity?
• When has reach become sufficient?

The Reach Design Template

Target learner: who needs the intervention?

Target capability: what must change?

Active ingredient: what cognitive operation must reach the learner?

Initial setting: where will it be built?

Support level: what help is allowed?

Representation set: which forms should be sampled?

Difficulty range: where is the current useful band?

Time reach: immediate, delayed, long-term?

Authentic context: where must it eventually work?

Minimum viable reach: what survives low-capacity weeks?

Expansion rule: what context comes next?

Stop rule: when can dedicated reach contract?

Worked Reach Design: Mathematics

Target: Mira.

Capability: quadratic method selection.

Active ingredient: discriminate before solving.

Initial: tuition, untimed mixed cases.

Support: structural cue only.

Representations: equations, graph-linked problems, short word context.

Difficulty: near current level.

Time: immediate then delayed.

Authentic: timed mixed paper.

Minimum: three classifications + one solve.

Expansion: fresh unfamiliar surface.

Stop: stable authentic mixed selection under delay and independence.

Worked Reach Design: English

Target: Jonas.

Capability: evidence-bounded inference.

Active ingredient: evidence selection + claim calibration.

Initial: narrative passage.

Support: evidence-location cue if needed.

Representations: narrative, article, speech.

Difficulty: familiar vocabulary first, then more complex texts.

Time: immediate and delayed fresh passage.

Authentic: timed comprehension.

Minimum: two evidence-to-claim comparisons.

Expansion: cross-genre.

Stop: calibrated claims across fresh genres independently.

Worked Reach Design: Science

Target: Nadia.

Capability: evidence-to-mechanism explanation.

Active ingredient: evidence → pattern → mechanism → bounded conclusion.

Initial: familiar experiment.

Support: terminology clarification allowed.

Representations: graph, table, apparatus diagram.

Difficulty: clean setup before noisy unfamiliar setup.

Time: immediate then spaced return.

Authentic: mixed examination application.

Minimum: one complete chain.

Expansion: new Science topic.

Stop: mechanism travels across topics and evidence forms.

Research Foundations and Evidence Boundaries

The strongest direct current practice anchor is AERO’s Staying on track: Monitoring implementation outcomes, published 15 September 2026, which explicitly identifies fidelity and reach as implementation outcomes to monitor. AERO’s Context first guide, published the same day, emphasises contextual tailoring rather than assuming one implementation form fits all settings.

EEF’s 2026 Guide to Inclusive Teaching provides a complementary educational frame by asking what adaptations or additional support some pupils may require and whether those supports help rather than hinder learning. Its relevance to reach is conceptual: equal access to teaching does not guarantee equivalent access to the learning operation.

The 2025 paper Developing Implementation & Scaling Capacity in Education documents how implementation capacity must be developed deliberately when educational approaches expand across complex systems. EEF’s ongoing evaluation of scale-up programmes is similarly examining processes, contextual factors, recruitment, fidelity and sustainment during scaling.

These sources operate mainly at school and system level. This article translates their implementation logic into learner-level and small-group training. It does not claim that a child’s home practice can be measured with the same constructs or methods as programme-level reach. The bounded claim is simpler: a training mechanism cannot influence a learner, context or performance condition that it never meaningfully reaches; implementation should therefore track more than attendance and material distribution when broader transfer is expected.

Final Principle

A learning system can look busy while the important mechanism barely reaches the learner.

A child can attend.

Complete.

Submit.

Score.

And still never practise the decision, retrieval, explanation or transfer that the future performance actually requires.

Reach asks us to follow the mechanism all the way from design to lived performance.

Did it reach the learner?

Did it reach the weak link?

Did it reach the relevant representations?

Did it reach independence?

Did it reach home and school?

Did it reach after delay?

Did it reach authentic conditions?

A training plan has meaningful reach when the right learning operation gets to the right learner, in the right conditions, often enough that the capability can eventually survive without the plan standing beside it.