Skip to main content
The state of ai impact assessment
Which Test Proves Bad News Reaches the Board?Enterprise Risk Management
5 min readFor Internal Auditors

Which Test Proves Bad News Reaches the Board?

By Governance AuthorityPublished

A risk can be technically monitored and still be effectively invisible if the people closest to it don't trust, or aren't able, to put the full picture in front of decision-makers. That is the governance question raised by allegations at the UK National Energy System Operator (NESO). An independent report by Eversheds Sutherland identified serious governance and culture failings said to be undermining transparency and oversight of critical electricity supplies. Separately, control-room whistleblowers accused management of covering up the risks to, and extent of, power outages during June's heat waves. Those accusations are allegations, not established findings. They still raise a concrete audit question: how will you test whether bad news reaches decision-makers?

The decision you are facing

You have limited audit capacity and one question to answer: does material operational risk information reach the board and its committees accurately and on time? You can't answer it by confirming that a risk register exists or that a whistleblowing policy has been approved. You need to choose a testing approach, and the right one depends on what you already suspect.

Key factors that affect your choice

Work through these before selecting a path.

  • Existing signals. Have speak-up reports, exit interviews, survey results or near-miss data suggested that frontline staff doubt the reporting chain? If yes, favor a culture-led approach.
  • Distance between the frontline and the board. The more management layers that summarize, filter or rate operational information, the greater the opportunity for distortion.
  • Incentive exposure. Where performance targets, regulatory relationships or reputational concerns reward reporting "no issues," bad news faces a structural headwind.
  • Independence. If internal audit reports through the same management line whose reporting is in question, an external review may be the credible option. Principle M of the UK Corporate Governance Code expects the board to establish formal and transparent arrangements for internal and external assurance, and Provision 29 requires the board to monitor the risk management and internal control framework and report on its effectiveness.
  • Speak-up credibility. Provision 6 of the Code expects the board to routinely review whistleblowing arrangements and ensure proportionate, independent investigation and follow-up. A channel that exists but is not trusted provides false comfort. The COSO ERM Framework emphasizes information, communication and reporting: risk information has to be relevant, timely and reliable to be useful.

Path A: Trace-back testing of the reporting chain

Choose this when you have no specific red flags but need baseline assurance, or when you want a repeatable annual test.

If your concern is whether risk information is distorted as it moves upward, then select a sample of operational events, such as near-misses, capacity shortfalls, alarms and degraded-service episodes. Start from raw frontline records, such as shift logs, control-room records and incident tickets. Trace each one forward through every summary, dashboard and committee paper. Compare the original severity and extent with what the board saw.

What you are testing:

  • Whether severity ratings change between levels, and who approved the change.
  • Whether events appear in executive reporting at all, and how quickly.
  • Whether aggregated metrics mask the extent of an issue that raw logs reveal.

Trace-back testing generates documentary evidence. Its limit is that it finds only what was recorded. If frontline staff stopped logging issues they believed would be ignored, the trail will look clean.

Path B: Culture-led evidence gathering

Choose this when signals suggest staff distrust the chain, or when Path A results look unusually tidy.

If you need to distinguish an isolated reporting failure from a wider culture problem, then combine independent evidence sources rather than relying on one:

  • Confidential interviews with frontline and supervisory staff, conducted without management present.
  • Analysis of speak-up data: volumes, source levels, time to triage, outcomes, and any pattern of reporters leaving or being reassigned afterward.
  • Comparison between what frontline staff say happens and what management attests.
  • Review of board and committee minutes for how challenge was received and recorded.

The pattern matters more than any single data point. One discrepancy in one team suggests a process failure. Similar discrepancies across teams, combined with low trust in the speak-up channel, suggest a cultural condition that process fixes alone won't cure.

If concerns conflict with management's account of operational risk, then the investigation should not be run by the function whose account is challenged. Route it to the audit committee chair or an independent investigator, protect the reporter's identity and employment position, and document how the conflict of accounts was resolved.

Path C: Commission an independent external review

Choose this when the allegations touch senior management, internal audit's own independence is open to question, or stakeholders and regulators will not accept an internally produced conclusion.

If you commission one, then set the terms carefully. Define the scope to cover governance and culture, not only policy compliance. Give the reviewers direct access to staff and records. Agree in advance who receives the report and how findings will be disclosed.

Be realistic about limits. An external review is a point-in-time assessment. It can diagnose, but it can't prove that anything has changed. The test of the review is what follows:

  • Named owners and dates for each recommendation.
  • Audit committee tracking of closure, with evidence rather than management assertion.
  • A follow-up reporting-chain test, using Path A methods, after remediation.
  • Changes to incentives and accountability where findings implicate them.
  • Reporting in the annual Governance Disclosure on how the board satisfied itself that the changes are working.

Summary matrix

Factor Path A: Trace-back testing Path B: Culture-led evidence Path C: Independent review
Best trigger Baseline assurance Signals of distrust Senior management implicated; independence in doubt
Primary evidence Documents and logs Interviews, speak-up data Full-access review
Main limitation Misses unrecorded issues Subjective; needs triangulation Point-in-time; needs follow-up
Independence level Internal audit Internal audit, ideally reporting to audit committee External
Code anchor Provision 29 Provision 6 Principle M
Proof of durability Repeat test Repeat survey and speak-up trend Tracked remediation and re-test

These paths aren't mutually exclusive. Many functions will run Path A annually, add Path B when signals appear, and recommend Path C when neither can be trusted to be independent. The principle holds across all three: in critical infrastructure, the integrity of risk reporting is itself a resilience control, and it should be tested like one. For your next audit plan, select a sample of operational events and trace each from the frontline record to the board report.

Promotional banner for the CUI buildout guide

You Might Also Like