The Training Provider Business Health Check: A Six-Area Audit for 2026/27

Last reviewed: 26 July 2026.

Why this review matters more in 2026/27

Providers run self-assessment on quality and prepare rigorously for inspection. Far fewer run a structured review of the business — the portfolio, the employer base, the margin, the concentration risk. In a stable funding environment that omission is survivable. 2026/27 is not a stable funding environment.

Four changes land close together:

• Funding is withdrawn from 16 apprenticeship standards on 1 September 2026, including Team leader, Operations manager and Coaching professional.
• Employer co-investment rises to 25%, a fivefold increase on the previous 5%.
• Levy funds now expire after 12 months rather than 24, pushing more levy employers into co-investment sooner.
• Responsibility for apprenticeships moved from DfE to DWP in April 2026, and the 2026/27 rules published under it tighten subcontracting and accountability.

Individually, each is a compliance item with an owner. Together they are a commercial stress test, and they interact: a provider heavy in management standards, serving mostly non-levy employers, with a thin margin and one dominant employer account, is exposed on four fronts at once. Tracked as four separate workstreams by four different people, that combined exposure is invisible until it arrives.

Area 1: Portfolio and funding exposure

Start here, because in 2026/27 this is where the risk concentrates.

• Starts by standard. What share of last year's starts sat on each standard? Anything above roughly 20% is a concentration.
• Defunding exposure. What share sat on the 16 standards losing funding in September, and what replaces that volume?
• Levy versus non-levy split. Non-levy volume carries the co-investment increase directly.
• Age profile. Funding policy is redirecting towards younger learners; a portfolio weighted to over-25 adult CPD is exposed to the direction of travel, not just the current rules.
• Funding band mix. Low-band, high-volume delivery behaves very differently under margin pressure than high-band, low-volume delivery. The funding band estimator is a quick way to build this picture across the portfolio.
• Sector alignment. How much of the portfolio sits in industrial strategy priority sectors?

The defunding round is the useful lesson here: a standard moved from core portfolio to unfundable with roughly five months of notice. The stated criteria — support for young people, industrial strategy alignment, forming a core part of a career pathway — apply to considerably more of the catalogue than the 16 named. Score every material standard in your portfolio against those three tests now, while diversifying is a choice rather than a reaction.

Area 2: Employer base

• Revenue concentration. What share comes from the largest employer, and the largest five? Above roughly 15% from one account warrants a named mitigation plan.
• Account health. Which employers have reduced starts year on year, and is anyone tracking that as a signal rather than a number?
• Levy position. Which levy employers are heading for an exhausted balance under 12-month expiry, and therefore into co-investment?
• Single-standard dependency. Which employers rely on a defunded standard as their only route? Those relationships lapse entirely unless something replaces it.
• Relationship depth. How many accounts rest on one named contact who could leave?
• Pipeline quality. Are planned starts committed, or hoped for?

The cross-cutting finding this area usually produces is that concentration is worse than assumed, because it compounds: the largest employer is frequently also on the most exposed standard. Our employer engagement guide covers the account structures that reduce this, and the levy account guide covers transfers as a co-investment mitigation.

Area 3: Delivery capacity and operations

• Caseload per tutor, and the spread rather than the average — averages hide the person carrying 60.
• Achievement and retention by standard, cohort and tutor.
• Gateway and EPA flow: where learners queue, and for how long.
• Off-the-job evidence: whether OTJ hours are captured as they happen or reconstructed under pressure.
• Administrative load: how much tutor time goes to recording rather than teaching.
• Data quality: whether ILR submissions require significant manual correction.
• Subcontracted delivery: volume, oversight, and exposure to the tightened DWP definitions.

Retention deserves particular attention in 2026/27. On a defunded standard, a withdrawn learner can no longer be re-recruited after September — the loss is permanent in a way it has not previously been. That materially raises the commercial value of retention work on affected cohorts.

Area 4: Financial visibility

• Margin by standard. Not average margin — margin per standard, fully loaded with tutor and administrative time.
• Cash-flow visibility against the funding calendar.
• Cost to serve per learner, and how it varies by employer.
• Break-even volume per standard, and how far current volume sits above it.
• Fixed cost base and how quickly it could flex if volume dropped.
• Non-funded income: what share of revenue does not depend on funding policy at all?

Margin by standard is the number most often missing and most often decisive. Providers frequently discover that a high-volume standard they consider core is close to break-even once tutor time is loaded properly, and that a smaller programme they nearly discontinued carries the business. You cannot make a sensible portfolio decision under funding pressure without it.

Area 5: People and leadership

• Tutor recruitment and retention, including time-to-hire for occupationally competent staff.
• Assessor and IQA capacity against forecast volume.
• Key person risk: which roles have no cover, and where knowledge lives only in one head.
• Leadership bandwidth for change on top of delivery.
• Succession for critical roles.
• Staff development routes now that Learning and skills assessor and mentor are among the defunded standards — providers using those to build internal capacity need a funded alternative or a budget line.

Area 6: Growth readiness

• Capacity to scale without proportional cost increase.
• New provision in development, and how quickly it could reach delivery.
• Units and short courses: whether you can deliver apprenticeship units, which are the funded replacement for the adult CPD volume the defunding removes.
• Commercial capability: whether you can price, contract and sell training directly, which determines whether defunded standards become lost revenue or retained revenue.
• Young-entrant provision: alignment with foundation apprenticeships and the hiring incentives.
• Systems headroom: whether your platform supports new provision types without a rebuild.

Commercial capability is the sharpest differentiator in this list. All 16 defunded standards remain deliverable to privately funded apprentices. Providers able to price and contract directly retain revenue that levy-dependent competitors simply lose.

Scoring and prioritising

Score each of the six areas 0 to 4 — absent, recognised, partial, established, managed — and, as with any diagnostic of this kind, resist the urge to average them. A provider strong in delivery and people but scoring 1 on portfolio exposure and financial visibility is not "mostly fine". It is a well-run operation with a concentration problem it cannot currently see.

Then sort findings into three buckets:

Anything that cannot be assigned an owner and a date belongs in "watch", not "build". A plan with fourteen priorities has none.

Running the review

Three days of structured work covers it for most providers.

• Day one: data extraction — starts by standard, employer revenue concentration, achievement and retention, margin by standard
• Day two: interviews with delivery, employer engagement, MIS and finance, conducted separately
• Day three: scoring workshop with the senior team, bucket allocation, owners and dates

Two things determine whether it produces anything useful. Extract the data before the workshop, because a session spent debating what the numbers are never reaches what they mean. And interview delivery staff separately from senior leadership — the gap between the two accounts of how something works is consistently one of the most informative findings.

The output should be a scored profile across the six areas, a named concentration risk per area with its mitigation, a prioritised list with owners and dates, and an explicit statement of what you have decided not to do this year.

Run it before budgets are set rather than after. A review that arrives once the plan is written becomes a document that justifies decisions already taken, which is a considerably less valuable thing than one that shapes them.