The Complete Guide to Apprenticeship Risk Tracking

Apprenticeship risk tracking should help a provider answer a practical question: which apprentices need support now, what is getting in their way and who is going to act?

A list of red, amber and green learners is only a starting point. Effective apprenticeship performance management connects timely data to an explanation, an intervention and a follow-up check. It also gives leaders a clearer view of programme performance and the factors shaping future achievement.

This guide covers withdrawal risk, learner retention, funded programme tracking and Qualification Achievement Rate (QAR), with a practical approach for operations, quality, compliance and MIS teams.

Scope: apprenticeship provision in England. Official guidance checked on 7 October 2026. The workflow recommendations can be adapted more widely, but QAR and funding requirements should be checked against the rules for the relevant nation, programme and reporting year.

What is apprenticeship risk tracking?

Apprenticeship risk tracking is the process of identifying barriers to progress or achievement, reviewing the evidence with the apprentice and relevant staff, and managing the support needed to address them.

It works at two levels. At learner level, it helps a coach decide what to do next. At programme level, it helps leaders identify recurring problems in delivery, employer involvement, assessment, support or capacity.

Keep three types of risk visible:

• Learner risk: barriers to engagement, progress, wellbeing or successful completion.
• Delivery risk: issues in staffing, curriculum, employer support or assessment arrangements that may affect several apprentices.
• Data and funding risk: incomplete evidence, inconsistent records or unresolved questions about the funding position.

These can overlap. A missing review record might reflect an administrative delay, a missed review or a learner who needs help. Check which explanation is supported before assigning an intervention.

How is QAR calculated for apprenticeships?

QAR = eligible achievers ÷ eligible leavers × 100.

The official QAR methodology for 2025 to 2026 uses ILR data, programme-aim matching, exclusions and a hybrid end year to determine the calculation. The hybrid year considers planned end, actual end, achievement and reporting years, subject to the detailed rules. Some overdue continuing records and breaks can be treated as leavers.

For the same eligible population, QAR retention is completers divided by leavers; pass rate is achievers divided by completers. QAR combines both elements. It is not achievers divided by starts, and its retention measure differs from the proportion of a starting cohort still participating.

Use the specification for the reporting year being analysed. Label internal forecasts separately from official results.

For management purposes, ask MIS to explain differences between your operational view and the official dataset at record level. Keep the extract date, source and calculation version alongside the result. A forecast is more useful when the team can see the assumptions behind it.

How do apprenticeship providers identify learners at risk of withdrawal?

Bring together changes in engagement, progress, employer circumstances and support needs. Treat these as prompts for investigation, rather than proof that an apprentice will withdraw.

The following signals provide a practical starting point for review:

Consider changes from the apprentice’s own pattern as well as agreed programme expectations. A single missed activity may have an understandable explanation; a combination of unresolved signals deserves attention.

Maintain an “insufficient information” state. Where a feed has not refreshed or activity is recorded elsewhere, investigate the data gap. Do not automatically describe the apprentice as disengaged.

Staff conversations remain essential. A learner may disclose a barrier that no activity record captures, while a dashboard flag may be explained by an agreed adjustment. Record the reason for the risk judgement and review it as circumstances change.

Safeguarding concerns require the provider’s safeguarding response immediately. Keep sensitive case information in the appropriate system and share only what staff need for their responsibilities.

Turn risk signals into an intervention workflow

A risk flag needs a route to action. Use a simple sequence that the team can follow consistently:

1. Validate: check the source, date and apprentice’s current circumstances.
2. Understand: discuss the barrier with the apprentice and involve the employer or support team where appropriate.
3. Agree: record the intended outcome, action owner and next review date.
4. Act: deliver the agreed support and record what happened.
5. Review: check whether the barrier has reduced and whether progress has resumed.
6. Escalate or close: arrange further support if needed, or record the evidence supporting closure.

Choose the intervention to match the barrier. More reminder messages will not resolve a lack of protected learning time. A revised workload will not resolve a technical access problem. The first useful action is often establishing what is preventing progress.

If you use risk categories, define the response attached to each one. Staff should know who reviews a new flag, how urgent cases are escalated and when an overdue action receives management attention. Agree these arrangements around your delivery model and capacity.

A practical apprenticeship risk register template

For each open risk, capture enough information for another authorised colleague to understand the situation and continue the work:

• Apprentice and programme: a controlled link to the correct source record.
• Signal and evidence: what changed, the source and when it was checked.
• Context and barrier: what the apprentice and relevant staff have confirmed.
• Risk judgement: category, rationale and any uncertainty.
• Agreed action: the support to provide and the outcome sought.
• Accountable owner: the person responsible for coordinating follow-through.
• Next review: date, expected evidence and escalation route.
• Outcome: what changed, what remains unresolved and the closure decision.

Link to existing records where possible. Avoid creating a parallel spreadsheet that becomes another version of the learner’s status. Keep access proportionate to the information involved.

How do apprenticeship providers track funded programme performance?

Use a connected view of delivery, outcomes, funding and data quality. A provider-level average should lead to the programme, employer or caseload needing attention, and then to the underlying records.

A useful management view includes:

• Participation: starts, continuing learners, agreed breaks and withdrawals, with clear definitions.
• Delivery: engagement, progress reviews, planned learning and unresolved support actions.
• Completion: approaching completion dates, assessment readiness, completed learning and confirmed achievements.
• Quality: sampling findings, learner and employer feedback, and recurring improvement themes.
• Funding: recorded funding, payment reconciliation, forecast assumptions and exceptions requiring review.
• Data confidence: refresh dates, missing records, conflicting statuses and reconciliation issues.

Show counts alongside percentages and make the population and period visible. A small programme can move sharply after a small number of outcomes. Compare like-for-like cohorts and explain changes in programme mix before drawing conclusions about performance.

Keep expected future income separate from recorded earnings and received payments. When learner circumstances change, ask the responsible team to assess the funding effect under the applicable rules; a risk flag alone does not establish a financial adjustment.

The apprenticeship accountability framework also provides an external reference for performance review. Use its current indicator definitions and thresholds. Its withdrawal measure uses a different population from the QAR leaver calculation, so the percentages are not interchangeable.

The January 2026 framework suspended the separate indicators for breaks, off-the-job training and end-point assessment organisation data. Those areas remain useful operational checks; distinguish your internal measures from the framework’s active indicators. Thresholds support risk identification and review, rather than guaranteeing automatic contractual action.

For the data foundations, see how to manage apprenticeship data across systems.

How can a training provider improve its QAR?

Start by separating the underlying problems. Apprentices leaving before completing their learning need a different response from apprentices who finish training but face unresolved assessment barriers.

Focus apprenticeship quality improvement on actions supported by evidence:

• Review early experience: check whether the programme, employer arrangements and initial support match the apprentice’s needs.
• Address recurring withdrawal reasons: use consistent categories, retain the individual context and investigate themes across programmes.
• Protect learning opportunities: resolve barriers to participation and agreed training activities with the apprentice and employer.
• Prepare for assessment: identify outstanding requirements early and coordinate support against the applicable assessment plan.
• Check delivery capacity: review whether caseloads and staffing allow agreed support and timely follow-up.
• Evaluate interventions: check which actions helped and where a different response is needed.

Record status and outcomes accurately. Leaving an inactive learner marked as continuing, or changing dates simply to improve a report, obscures the problem. Learner retention should mean sustained, worthwhile participation with appropriate support.

Prioritise support according to learner need and urgency. Scenario modelling can help leaders understand potential performance effects, but an apprentice’s access to help should not depend on how much their outcome would move a headline rate.

Use forecasts to expose assumptions

Build a forecast from the records and assumptions your team can explain. Show what is confirmed, what is expected and what remains uncertain.

Separate scenarios such as expected achievements, delayed outcomes and possible non-achievement. Explain which records move between scenarios and why. Where timing changes, review both the achievement assumption and the population used in the calculation.

An automated risk score should remain a prioritisation aid. Unless its probabilities have been tested against relevant outcomes, do not present a score as a proven probability of withdrawal. Allow staff to challenge it and record their reasoning.

Review missed risks as well as unnecessary alerts. A system that produces many flags may consume staff capacity without identifying the apprentices who most need support.

Review at a pace that allows action

Agree a review rhythm around the decisions being made. As a practical starting point, coaches can review new exceptions during their normal working day, managers can check open interventions weekly, and leadership teams can examine programme trends monthly.

At each review, ask: what changed, who acted, what was the result and what needs escalating? Track the age of unresolved risks and overdue actions alongside the total number flagged.

Use a consistent follow-up period when comparing intervention outcomes. A learner improving after support is useful evidence, but does not by itself prove that the intervention caused the improvement. Combine outcome tracking with case review and learner feedback.

How AiVII supports apprenticeship performance management

AiVII’s QAR dashboard supports achievement tracking, cohort analysis, forecasting and scenario modelling. Its wider reporting and analytics brings performance information into views used by provider leaders and delivery teams.

Use those views to identify questions, trace them to the underlying records and agree action with the responsible staff. The management value comes from combining visibility with an owned support process and a check on the result.

Want to see how this works for your provision? Explore AiVII’s QAR dashboard and discuss how your team can connect risk tracking, learner support and programme oversight.

Frequently asked questions

Which data should we start with?

Start with reliable learner status, planned activity, recent meaningful contact, progress and open support actions. Add further signals when the team can explain their source and act on them.

Does being past the planned end date mean an apprentice will withdraw?

No. Investigate the reason, distinguish ongoing training from assessment-stage delays and agree a realistic action. An overdue date is a question to resolve, rather than a confirmed outcome.

Can a break in learning be managed as a support measure?

Review the apprentice’s circumstances under the applicable funding rules. Where a break is appropriate, maintain the required records and a clear contact and return plan. Do not use a break simply to remove an exception from a dashboard.

How do we know whether risk tracking is helping?

Check whether validated concerns receive timely action, whether barriers are resolved and whether participation and progress improve. Review learner feedback, repeated risks and missed cases alongside headline outcomes.