Why do operators fail the Earned Recognition audit?

Most failures are evidence failures, not compliance failures — the work was done, but the record that proves it is incomplete, unsigned, or impossible to produce on demand. The audit measures the evidence, so a well-run fleet with poor record-keeping can still fail.

What's the difference between a process failure and an evidence failure?

A process failure means the operation isn't doing something — e.g. brake tests aren't frequent enough — and needs an operational change. An evidence failure means the work happens but the record is missing, unsigned or unretrievable, which is a faster, cheaper fix. Most ER failures are evidence failures.

What is the most common Earned Recognition audit failure?

Incomplete or unsigned safety inspection records. Because the maintenance KPI targets are 100%, a single missing record or blank signature box is a failure — and it's almost never because the inspection didn't happen, but because the proof wasn't captured to standard.

What is an open defect loop and why does it fail an audit?

An open defect loop is a reported defect with no record of its closure — who fixed it, when, with what evidence — or a defect that's genuinely still unrepaired. It maps to the 100%-target 'driver-reported defects actioned' KPI, and auditors are trained to look for it as a sign of a system that captures problems but doesn't close them.

Can slow record retrieval fail an audit?

Yes. When an auditor samples vehicles' records they expect a complete set in minutes. If assembling it means searching filing cabinets and phoning the maintenance contractor, the slow retrieval is itself a finding — it signals a system held together by individual knowledge rather than structure, even when the records genuinely exist.

Why do policies fail the audit even when they exist?

Because the additional-policies section rewards policies that are lived, not just written. A stack of freshly-dated policies no staff have seen is transparent to an auditor, who interviews drivers and cross-checks. The fix is reviewing policies on a real schedule, re-issuing them, and keeping acknowledgement records.

How do agency drivers and hired vehicles cause failures?

They're the operation's edges, where control is easiest to lose under time pressure — an agency driver who started before their licence and CPC were checked, or a hired vehicle put into service before maintenance verification. The compliance is usually achievable; what's missing is a standing process that captures it every time.

How can I find failures before the auditor does?

Self-diagnose: pull five random vehicles' records and time it; trace ten recent defects from report to sign-off; check the planner runs six months ahead with no slipped intervals; confirm your tachograph cadence against the 14/42/7/28 standard; and ask a driver about a policy. Every failure an auditor could find, you can find first.

How do digital records reduce audit failures?

They eliminate the entire evidence-failure class. A digital walkaround check can't be submitted unsigned or undated, a closed-loop defect can't quietly stay open, and 15 months of records can't become unretrievable — so the audit tests your compliance rather than your record-keeping.

Earned RecognitionAudits·By Andrei Conea, Founder·9 July 2026·12 min read

Common DVSA Earned Recognition Audit Failures

Q: What is the root cause behind most audit failures?

An optimistic self-assessment. Because the audit is against published standards, a genuine surprise usually means the operator scored themselves against the process they intended to have rather than the evidence they could produce. An honest self-assessment surfaces every failure mode months in advance.

Why operators fail the DVSA Earned Recognition audit — the recurring failure modes, the root cause behind them, and how to self-diagnose before the auditor does.

Here is the uncomfortable truth about the DVSA Earned Recognition audit: most operators who fail it are not badly run. They maintain their vehicles, their drivers are within their hours, their defects get fixed. They fail because the evidence of all that is incomplete, inconsistent or unretrievable — and the audit measures the evidence, not the intention.

That distinction is the single most useful thing to understand before an audit, because it changes what you fix. This guide works through the failure modes DVSA-authorised auditors see again and again, explains the root cause behind each, and — crucially — separates the handful that are genuine compliance problems from the majority that are evidence problems with a much cheaper fix. It builds on the full audit guide and DVSA's published scheme guidance.

Process failures vs evidence failures

Most Earned Recognition audit failures are evidence failures, not compliance failures: the work happened, but the record that proves it is missing, incomplete, unsigned, or impossible to produce on demand.

This matters because the two have completely different fixes. A genuine process failure — you're not doing brake tests often enough — requires changing the operation. An evidence failure — you do the brake tests but the records are scattered and half-unsigned — requires changing how you capture and store records, which is faster, cheaper, and where the majority of failures actually sit. Read every failure mode below through that lens: for most of them, the operation is fine and the record is the problem.

The incomplete or unsigned record

The most common failure, and the purest evidence failure. The maintenance KPIs demand a complete set of safety inspection records, correctly completed and signed off, at 100% — so a single inspection with no record, or a record that was never signed by a competent person, is a miss. Not a small deduction: a failure against a 100% target.

Auditors see this constantly on paper-based fleets, and it is almost never because the inspection didn't happen. It's because a sheet went missing, a signature box was left blank, a date was ambiguous, or the record sat in a maintenance contractor's office and never made it back. The work was done; the proof wasn't captured to standard. The fix isn't "do more inspections" — it's "make it impossible to produce an incomplete record," which is exactly what a digital system does by construction. Understand how this maps to the numbers in the KPIs explained.

The open defect loop

The second most common, and it maps directly to the driver-reported defects actioned KPI (again, a 100% target). A driver reports a defect; it gets fixed; but nothing records the closure — who repaired it, when, with what evidence. Or worse, the defect is genuinely still open, sitting on a list nobody worked through.

Auditors are trained to look for exactly this: a reported defect with no visible rectification is one of the clearest signs of a system that captures problems but doesn't close them. The root cause is almost always that defect tracking depends on a person remembering to chase, rather than a system that won't let a defect disappear. Closed-loop defect management turns "did we fix that?" from a question into a tracked status — the difference between an audit finding and a clean record.

Forward planning that ran out

A genuine process failure this time. The vehicle standards section expects a forward maintenance planner covering at least six months ahead, with inspections happening at the frequency you declared. Two versions of this fail:

The planner only runs a few weeks ahead, so there's no evidence of systematic forward planning.
The plan exists, but in a busy period an inspection interval was quietly stretched — a six-weekly inspection that happened at eight weeks.

The second is the more damaging because it shows the frequency isn't actually controlled. The fix is to build the planner out to six months, reconcile the last six months of planned-versus-actual, and treat a slipping interval as an incident to investigate rather than a scheduling inconvenience.

Tachograph cadence failures

The driver-side equivalent of the record problem. The audit expects driver cards downloaded at least every 14 days, vehicle units every 42, analysis within 7 days, and infringements debriefed within 28. Fail the cadence and two things happen: infringements accumulate undetected, pushing you over the drivers' hours KPI thresholds, and you can't evidence that you were managing them.

The root cause is treating tachograph analysis as a monthly or quarterly catch-up rather than a continuous process. By the time a quarterly download reveals a pattern of infringements, the KPI period has already recorded them and the debrief window has closed. Operators who run the cadence weekly catch problems inside the period and correct them; operators who batch it fail on both the number and the evidence of managing it.

Records that can't be produced

A failure that has nothing to do with whether you're compliant and everything to do with whether you can prove it quickly. When an auditor samples five or ten vehicles' records, the expectation is that a complete set appears in minutes. If assembling it means searching filing cabinets across depots and phoning the maintenance contractor, the sampling grinds — and slow retrieval is itself a finding, because it signals a system held together by individual knowledge rather than structure.

This one frustrates good operators most, because the records genuinely exist. But an auditor can only assess what you can put in front of them, and an afternoon of searching reads as a weak system regardless of the underlying compliance. Fifteen months of records that export in one click removes the failure mode entirely.

Policies that aren't lived

The additional-policies section is where last-minute preparation is most visible. A stack of freshly-dated policies — road safety, drink and drugs, mobile phones, fatigue, whistleblowing — that no member of staff has seen is transparent to an experienced auditor, who will interview drivers and cross-check. The failure isn't the absence of a policy; it's a policy with no evidence it's been communicated or applied.

The fix is unglamorous: review policies on a real schedule, re-issue them to staff, and keep the acknowledgement records. A policy is only worth the evidence that people follow it.

The unmanaged edges

The last cluster is the operation's boundaries, where control is easiest to lose: agency drivers who started before their licence and CPC were checked, hired vehicles put into service before maintenance was verified, sub-contractors nobody vetted. The operation-management section examines all of these, and they fail because the edges are handled ad hoc — a driver needed today, a vehicle hired in a hurry — rather than through a standing process with records.

The fix is a documented pre-start check for every agency driver, a maintenance verification before any hired vehicle enters service, and an approved-contractor list with the vetting to back it. The compliance is usually achievable; what's missing is the process that captures it every time, under pressure.

The failure behind the failures: a dishonest self-assessment

Nearly every failure above shares a root cause: the self-assessment was optimistic. Because the audit is conducted against published standards, a genuine surprise on audit day almost always means the operator scored themselves against the process they intended to have rather than the evidence they could produce. An honest self-assessment — scoring only what you could put in front of an auditor today — surfaces every one of these failure modes months in advance, while there's still time to fix the record base before it's sampled.

How to self-diagnose before the auditor does

Run this diagnostic against your own operation:

Pull five random vehicles' complete records and time it. More than a few minutes, or any gap, and you have the retrieval and completeness failures.
Trace ten recent defects from report to sign-off. Any that don't close cleanly are open-loop failures waiting to be found.
Check your planner runs six months ahead and that no interval slipped in the last six months.
Confirm your tachograph cadence against the 14/42/7/28 standard for the last quarter.
Ask a driver about a policy. If they've never seen it, it isn't lived.

Every failure an auditor could find is one you can find first — the only question is whether you look.

Eliminating the evidence-failure class

Group the failures by type and a pattern emerges: the process failures (forward planning, tachograph cadence, unmanaged edges) need operational fixes, but the evidence failures (incomplete records, open loops, slow retrieval, unlived policies) are all the same problem wearing different clothes — a record base that depends on people remembering to capture, sign, chase and file.

That entire class dissolves with the right daily systems. A digital walkaround check can't be submitted unsigned or undated. A closed-loop defect can't quietly stay open. Fifteen months of records can't become unretrievable. This is the honest value of a tool like HaulGuard in an Earned Recognition context — it doesn't fix a genuinely non-compliant operation, and it doesn't report your KPIs to DVSA (that needs a DVSA-validated system, whose status you should verify against DVSA's lists). What it does is make the most common failure modes structurally impossible, so the audit tests your compliance rather than your record-keeping.

The audit rewards the operation you already run

The reassuring conclusion beneath all of this: if you run a genuinely compliant operation, the Earned Recognition audit is not trying to catch you out. It's trying to confirm what's already true — and it fails you only when it can't, because the evidence isn't there to confirm it. Fix the evidence base, run an honest self-assessment, and the failure modes above stop being risks and become a checklist you've already cleared.

Start with the complete audit guide for the full picture, use the preparation playbook to work through the run-up, and if your records are still on paper, understand that it's not a filing preference — it's the source of most of the failures above. A two-minute digital walkaround is where a clean, complete, retrievable record base begins.