Who submits and the two deadlines
The structured programme concerns PAs with Partially satisfactory or Not satisfactory PMP outcomes and entities whose QC reviews produced those outcomes because of non-compliance. As part of PAOC sanctions, RCA and remediation plans are submitted within one month of outcome notification. Implementation must finish within twelve months, with critical areas addressed as soon as possible. ACRA may request revised analysis if it lacks robustness or appropriateness. These deadlines are distinct from the document’s suggested analytical methods, which are non-authoritative approaches rather than mandatory formulas.
Plan the exercise and appoint capable objective reviewers
Start timetabling as soon as findings are known or the order arrives, leaving time for analysis and action design. Reviewers need experience, technical skill, objectivity and authority. They may come from quality, methodology, training or monitoring functions but must consider whether their own systems caused the problem. Smaller firms without a suitable objective person may use external help; involving the appointed hot reviewer is encouraged where relevant. The role is not limited to a prescribed job title, and selecting an internal reviewer does not remove the need to challenge their own process.
Stage A: understand the gap and the quality-system deficiency
Share findings and compare actual work with required or expected procedures. For SoQM, the footnote identifies deficiencies where a required quality objective is absent, risks are unidentified or improperly assessed, responses fail to reduce risk through design/implementation/operation, or another required aspect is missing or ineffective. Knowing the precise gap helps identify necessary additional work. A broad finding title alone does not establish whether the issue is missing procedure, weak evidence, flawed design or ineffective operation.
Stage B: gather engagement and system evidence
For audit findings, gather management and reviewer experience, tenure and past results; team composition, hours, partner/EQCR involvement and specialists; assessed risks and planned procedures; and client industry, location, finances, tenure, fees and restatements. Review workpapers, risk planning, consultation notes and the guidance or templates actually used. For QC findings, understand design, implementation and operation through process owners and control operators, including their experience, environment and resources. The source distinguishes operationally accountable process owners across eight SoQM components from people performing controls. Evidence collection should reflect that distinction.
Interview people separately and seek facts before conclusions
Include partner, EQCR, managers, junior staff and relevant specialists, and where useful independence, technical or risk functions. Timely one-to-one interviews can avoid a group mindset and should use questions tailored to files and roles. Explain that the exercise is not performance appraisal or blame. Ask what happened and how, use open questions and probing follow-ups, avoid leading or negative wording, establish policy understanding and allow time to elaborate. Interview fact gathering is distinct from the later why-based causal analysis; the two recommendations are not contradictory.
Stage C: identify all contributing causal factors
Analyse patterns, anomalies and connections to identify immediate contributing factors. Human factors include skill, experience, workload, communication and judgement. Procedural factors include unclear or absent instructions and non-compliance. Organisational factors include budgets, staffing, leadership, supervision, accountability and policy design or communication. These factors bridge evidence and underlying causes. There is often no single controllable cause, so premature narrowing can leave actions incomplete. The categories are examples, not a closed checklist of permissible causes.
Stage D: determine deeper causes without a fixed question count
Apply professional judgement to themes and fundamental factors. Five Whys asks why repeatedly until further questioning adds no insight, using evidence; five is not a mandatory stopping point. Fishbone analysis maps the issue, major categories and contributing factors, looking for common factors across branches. Avoid five pitfalls: stopping at the first apparent cause, mistaking symptoms for causes, blaming individuals, inadequate data and rushed quick fixes. The output needs a defensible conclusion about underlying causes, rather than a diagram with no decision.
Stage E: make each action SMART and properly supported
Actions should address each cause with responsible people, resources and steps. SMART means specific, measurable, attainable, relevant and time-bound: define action and ownership, completion measures, feasible resources, cause relevance, phase dates, overall deadline, milestones and effectiveness testing. Leaders, teams and control owners must support changed systems and behaviours. Training can help but is not a cure for every finding. Avoid action without analysis, one solution for all issues, inadequate resources and missing follow-through. Regular progress reporting should handle changed conditions, not merely certify a plan has been written.
Assess whether past audits also need remediation
Prospective work prevents recurrence; retrospective work looks back and may require revised or restated information. Consider whether omitted procedures or evidence significantly affect support for the original report, whether pervasive material misstatements affect the true-and-fair view, and whether regulatory breaches have legal implications. The guidance poses considerations, not an automatic conclusion that every finding requires restatement or that all remediation can be future-only. The accountant must assess the facts and applicable obligations.
Stage F: implement, test and monitor
Communicate roles and deadlines, train for new procedures, document policy/process/system changes and update manuals, track status and challenges, test during implementation and adjust, then schedule ongoing monitoring with indicators where possible. Completing an action is distinct from testing that it works. ACRA’s review asks questions and comments on submissions; it is not a review of the firm’s SoQM and does not endorse or approve the analysis, plan or effectiveness. Those limits remain even where ACRA has not asked for another revision.
Appendix 1: five interview-question styles
Open questions invite a procedure or event walkthrough; probing questions explore missing resources, unclear steps and how frequency or severity evolved; clarifying questions check understanding; reflective questions ask about prevention and missed warning signs; knowledge questions test familiarity with policies, estimates or confirmation steps. These are non-exhaustive styles to tailor to the interviewee and evidence. Reflection is not an invitation to substitute speculation for the factual record, and knowledge testing within RCA is not the same as disciplinary appraisal.
Appendix 2: the complete potential-cause groups
Resource examples are preparer/reviewer skill gaps, missing industry or technical experience, time and budget pressure, understaffing, unclear responsibilities, excessive workloads and unsuitable team mix. Personal or ethical examples are weak work ethics, lack of scepticism, unwillingness to learn and deadline-driven corner-cutting. Process examples are outdated, inadequate or undocumented policy, weak review/supervision, failure to consult, poor project management, unclear accountability, insufficient training or partner involvement and late review. The list informs investigation rather than proving a cause without evidence. More than one category may explain the same finding.
Official source
This article independently explains the substantive contents of the official PDF, including the relevant conditions, procedures and annexes. The linked document remains the authoritative source for its original wording, and later changes should be checked separately.
Read the official PDF ↗