CLEAR Thinking
Design and Implementation Guide

Scaffolding CLEAR Thinking Within the Flow of Audit Work

A field manual for people who design the language, structures, workflows, interactions and conditions under which audit thinking happens.

Version 1.1  •  11 Aug 2026

About this guide

This guide is for people who design the environments in which financial-audit thinking takes place: audit tools and templates, systems and digital workflows, methodology and guidance, briefing and coaching practices, review approaches, learning events, leadership routines and AI-enabled support. Its subject is CLEAR Thinking — five linked cognitive moves (Clarify, Look Beyond, Evidence, Ask, Record and Reflect) — and how to scaffold those moves within the flow of real audit work, proportionately, without creating a parallel process or a documentation burden.

It is a design guide, not audit methodology. It does not replace auditing standards, prescribe accounting conclusions or mandate one template for all situations. Where the boundary between design guidance and methodology is unclear, the text says so and marks the point for local decision.

The CLEAR framework and the underlying five-move formulation originate in BDO exemplar research; this guide absorbs that content into a design system. Design interventions recommended here are hypotheses to be tested in realistic audit settings and updated accordingly.

The parts of the guide

Part 1 of this guide provides an in-depth description of CLEAR-Thinking, explaining how it functions as a system of thought to enable high-quality audits. The tables are presented in layers and should be used in sequence, then as key reference material. 

Part 2 provides a design system for embedding CLEAR-Thinking in the flow of work. Its principles and structure can be applied to the design of any touchpoint. 

Part 3 should be treated as a living section, updated as the design team learns what works and what doesn’t. It contains summary information for each element of CLEAR, including definitions, language patterns, coaching and review behaviours, what good looks like, and examples. These should all be treated as starting definitions and updated as lessons are learnt throughout the design cycle. As such, Part 3 should act as a primary design reference. 

Appendix: Activating thinking through language – design risks. This section outlines the two main language traps to avoid during detailed design. It should be read, understood, and used as a reminder when undertaking detailed design.

This guide is version 1.0. Its approach, principles and descriptions should be updated as the design teams’ experience grows and as learning emerges from applying CLEAR-Thinking in practice.

Contents

Part 1: Orientation

1.1. Background – the problem and solution

                  1.1.1 Disconnected thinking

                  1.1.2 Connected thinking

1.2.  CLEAR-Thinking defined

                  1.2.1 The five elements of CLEAR-Thinking

                  1.2.2 How the elements work as a system

                  1.2.3 Pseudo clear thinking

                  1.2.4 The role of leadership and culture

1.3.  The role of scaffolding

                  1.3.1 The environment already scaffolds thinking

                  1.3.3 Making scaffolding deliberate

                  1.3.4 Limitations and a caution

1.4.  The role of Meta-cognition (the skills required)

1.4.1 In design

1.4.2 In Coaching

1.4.3 In Practice

Part 2: Designing for CLEAR-Thinking

2.1 Experience-Led Design

                 2.1.1 Introduction

                 2.1.2 The Experience-Led Design process

2.2 Diagnose – identify critical intervention points

2.2.1 Experience the work from the auditor’s perspective

2.2.2 Identify opportunities to support CLEAR-Thinking

2.2.3 Prioritise critical intervention points

2.3 Define – specify the thinking required

                  2.3.1 Use CLEAR to help identify the required cognitive move(s)

                  2.3.2 Define the target thinking

                  2.3.3 Keep the thinking connected to reality

2.4 Design — Influence and activate the thinking

                  2.4.1 Identify and consider the effect of other relevant touchpoints

                  2.4.2 Five-step detailed design process

                  2.4.3 Design target

                  2.4.4 Evaluate the current experience

                  2.4.5 Create and refine the intervention

                2.4.6 Test the thinking produced

                2.4.7 Embed, learn and adapt

Page

1

1

2

3

3

5

6

6

7

8

8

8

8

9

9

9

12

12

13

14

14

14

15

16

16

16

17

18

18

18

19

19

19

20

22

Part 3: CLEAR -Thinking Element Descriptions for Design Reference

3.1 Connected thinking                

3.2  Clarify

3.3  Look Beyond

3.4  Evidence

3.5  Ask

2.7 Reflect & Report

25

26

28

30

32

34

Appendix: Activating thinking through language – design risks                                      38

Supporting communications materials

Part 1

Orientation

1.1     Background -  the problem and solution

This section offers a brief overview of the research findings that informed the development of the CLEAR-Thinking framework. It is not intended to present the research in full. Instead, it focuses on the key findings on the underlying causes of both strong and weak audit quality outcomes.

1.1.1 Disconnected thinking

The research behind this programme revealed a common root cause of many audit quality issues: disconnected thinking — numbers, schedules and analyses becoming detached from the real-world business events, people and processes that produced them. This is the same weakness that regulators, quality reviewers and experienced partners are regularly identifying: work that is procedurally complete, yet disconnected from the purpose, the risk being assured, and the audit business reality behind the numbers.

The table below highlights what this disconnect can look like in practice:

The disconnect

What it looks like

Purpose–procedure

The auditor completes the steps but cannot say what the work must establish, or what is still unresolved.

Number–reality

The figure is treated as an abstract amount, not a claim about customers, goods, contracts, decisions and conditions.

Evidence–claim

Evidence is gathered without distinguishing what it establishes, what it merely supports, and what lies outside its scope.

Finding–implication

An exception is cleared locally without considering related workstreams, or the overall judgement.

Evidence–updating

New information is added to the file but changes nothing — not the explanation, the work, the confidence or the conclusion.

Reasoning–conclusion

A conclusion cannot be reconstructed from the decisive facts, evidence and assumptions.

1.1.2  Connected thinking

By contrast, the research found that when BDO is at its best — delivering strong quality outcomes — connected thinking is at work. Our highest-performing auditors build a picture in the mind's eye of how the numbers connect to, and tell the story of, the real-world events that produced them. They understand the purpose of their work, what uncertainty is being addressed, what is being assured, and move fluently between the technical layer — standards, models, tests — and business reality: the things, processes, events, people and decisions behind the numbers. This high-performance thinking has now been codified in the CLEAR-Thinking framework.

The table below contrasts connected versus disconnected thinking in practice:

The connection

Disconnected thinking

Connected thinking

Purpose–procedure

The auditor completes the steps but cannot say what the work must establish, or what is still unresolved.

The auditor can state what the work must establish, why it matters to the audit, and what uncertainty remains unresolved once the steps are done.

Number–reality

The figure is treated as an abstract amount, not a claim about customers, goods, contracts, decisions and conditions.

The figure is translated into the real customers, transactions, assets, decisions and conditions behind it. The auditor asks what must have happened for the number to make sense.

Evidence–claim

Evidence is gathered without distinguishing what it establishes, what it merely supports, and what lies outside its scope.

Evidence is selected and weighed knowing exactly what it establishes between alternative explanations, what it merely supports, what questions it cannot answer, and whether it distinguishes between plausible explanations.

Finding–implication

An exception is cleared locally without considering related workstreams, or the overall judgement.

An exception is traced outwards — to related balances, assertions and workstreams, and considers whether it changes the risk assessment, planned work or engagement-level judgement— before it is cleared.

Evidence–updating

New information is added to the file but changes nothing — not the explanation, the work, the confidence or the conclusion.

New information is compared with what was expected; where it does not fit, the explanation, the work, the confidence or the conclusion is revised.

Reasoning–conclusion

A conclusion cannot be reconstructed from the decisive facts, evidence and assumptions.

Another informed person can reconstruct how the question, plausible explanations, decisive evidence, assumptions and residual uncertainty led to the conclusion.

1.2     CLEAR-Thinking Defined

It’s important to note that CLEAR is a system of thinking by design.  It is a system that works inside and outside of audit engagements – its repetition across all elements of an auditor’s role means it will become a mindset ‘just the way we do things around here’. Its elements work together, iterating and building on one another to develop sound, defensible conclusions. No element does its job in isolation. CLEAR thinking starts in planning and runs through to completion of the file.  It’s iterative – as you learn new information, you return to early stages: does the question need updating, do alternatives recede, or new ones emerge, are there new questions to ask? Whilst reviewers can use the CLEAR-Thinking framework to test the thinking on a file, it shouldn’t be applied retrospectively as a checklist. Thinking retrofitted at review is thinking that was missing when the work was performed, leading to low quality and significant rework.

1.2.1  The five elements of CLEAR-Thinking

The CLEAR-Thinking framework is a system of thinking, derived from research into BDO at its best. The table below briefly introduces each of the five elements of the framework and the question it keeps in front of mind. You can use the table as a quick-reference reminder—each element is explored in more depth in section 2 of this guide.

Element

Description

Key Question(s)

C-Clarify

Start with precision and purpose, connected to business reality. Define exactly what you are trying to do, understand, assess or conclude — and why it matters. Clarifying the purpose upfront helps focus the work on the right risk, outcome or decision, rather than simply completing a procedure.

“What exactly are we trying to do, understand or conclude — and why is that important?”

L- Look Beyond

Look past the obvious answer. Step back from the initial explanation and actively consider whether there are other plausible alternatives, risks or interpretations.

“What else could explain this — and what could we be missing?”

E- Evidence

Anchor in facts, not assumptions. Identify the evidence that matters and test whether it supports or contradicts the conclusion. Where evidence contradicts follow that line of thinking to an outcome – does it change the conclusion?

“What supports this?” and “what have I seen which may indicate that a different answer could be the right one?”

A- Ask

Drive better thinking through questioning. Use open, probing questions to test assumptions, methods and conclusions.

“Have we asked the right questions?”

R- Record and Reflect

Make thinking visible — and improve it. Record the conclusion clearly, explain why it is supported and how you considered and concluded on the potential alternative outcomes.  Then reflect and assess whether the reasoning remains coherent as a whole and defensible when viewed alongside the wider audit findings.

“Does this stand up to scrutiny — why do we stand behind it?”

The landscape table on the following page shows how each element of the CLEAR-Thinking framework—derived from the practices of high-performing auditors—directly addresses the underlying problems identified through root cause analysis.

CLEAR

Failure it primarily addresses

What happens when the element is missing

How the CLEAR element addresses the failure

Outcome

Clarify

Purpose–procedure
Number–reality

disconnects

The auditor begins with an assigned procedure, rolled-forward workpaper or account balance without being clear about what the work must establish. The number is treated as an abstract amount rather than a claim about real customers, goods, contracts, decisions, events or conditions.

Clarify defines the precise question, judgement or conclusion the work must support, why it matters and what business reality underlies it. It connects the procedure to the risk, assertion, decision and real-world claim being examined.

Work starts with a clear purpose. The auditor knows what must be understood, tested and concluded, and can distinguish relevant work from a tick box exercise that does not answer the question.

Look Beyond

Finding–implicationdisconnect

Premature commitment to one explanation

The first plausible explanation becomes the working answer. Exceptions are resolved locally, alternative explanations are considered only after commitment, and wider consequences for other balances, processes or judgements are missed.

Look Beyond requires the auditor to step back from the initial explanation, consider credible alternatives and ask what else could be happening. It extends attention beyond the immediate account or finding to connected risks, workstreams and engagement-level implications.

The auditor develops a broader and less anchored view of the situation. Alternative explanations are considered early, implications are traced across the audit, and important risks are less likely to remain hidden.

Evidence

Evidence–claim

Evidence–updating
disconnects

Evidence is accumulated because it is available or expected, without clarity about what it actually establishes. Contradictory information may be filed but does not change the explanation, planned work, confidence or conclusion.

Evidence connects each piece of information to the claim being tested. The auditor distinguishes evidence that confirms, contradicts, discriminates between explanations or lies outside the evidence’s scope, and follows contradictory evidence through to its implications.

Evidence becomes a basis for judgement rather than a collection of documents. The auditor can explain what the evidence establishes, what remains uncertain and how new information changed—or did not change—the emerging conclusion.

Ask

Unexamined assumptionsunderlying the purpose–procedure 

Evidence–claim
Finding–implicationdisconnects

Questions are limited to obtaining information or completing the work programme. Management’s explanation, the audit approach or the emerging conclusion is not sufficiently probed, and ambiguity remains concealed.

Ask uses open, probing and discriminating questions to test assumptions, explanations, methods and conclusions. It helps expose missing information, weak logic, contradictions and questions that the current work has not yet answered.

The audit moves from information collection to active inquiry underpinned by professional scepticism. Assumptions become visible, explanations are tested rather than accepted, and unresolved matters are identified early enough to influence the work.

Record & Reflect

Reasoning–conclusionWorkstream–judgement

Unresolved evidence–updating disconnects

The file contains procedures and evidence, but the conclusion cannot be reconstructed from the decisive facts, assumptions and judgements. Findings from separate workstreams remain fragmented, and documentation may rationalise a conclusion after the event.

Record and Reflect makes the reasoning visible: the conclusion, the evidence supporting it, the alternatives considered, the implications of contradictory information and the reasons for the final judgement. Reflection then tests whether the reasoning remains coherent when viewed within the audit as a whole, and if it raises implications for other elements in the audit/file.

A reviewer can follow how the auditor moved from the question, through alternatives and evidence, to the conclusion. Separate findings are integrated into the overall judgement, and audit file, and weaknesses or unresolved contradictions are more likely to be identified before the work is finalised.

1.2.2  How the elements work as a system

The table below describes how CLEAR works as a system of thinking. Its elements work together, iterating and building on one another to develop sound, defensible conclusions. No element does its job in isolation

CLEAR

Its distinctive contribution

What it provides to the other elements

How the elements work together

Clarify

Establishes the purpose of the work, the question to be answered, the real-world claim underlying the number or treatment, why it matters and what the work must establish.

Gives Look Beyond a clear issue around which to generate alternatives. Gives Evidence a defined claim to test. Gives Ask a purpose for its questions. Gives Record and Reflect a question and outcome against which completeness can be judged.

Clarify provides the organising question for the work, but that question remains provisional. Alternatives, questioning and evidence may reveal that the original framing was incomplete or wrong, requiring the auditor to clarify again.

Look Beyond

Prevents premature closure by generating credible alternative explanations, outcomes, risks and implications before the auditor becomes committed to one answer.

Gives Evidence competing explanations to discriminate between. Gives Ask lines of inquiry to pursue. Helps Clarify identify hidden assumptions or an overly narrow question. Gives Record and Reflect the basis for explaining why one conclusion was preferred over others.

Look Beyond widens the auditor’s field of view before evidence is interpreted. As questions are asked and evidence is gathered, some alternatives are rejected, others strengthened and new ones may emerge.

Evidence

Connects claims and explanations to facts. Determines what each piece of evidence establishes, what it contradicts, what lies outside its scope and how it changes the auditor’s confidence or conclusion.

Gives Clarify feedback on whether the original question was correctly framed. Narrows or expands the alternatives considered under Look Beyond. Provides the factual basis for further Ask questions. Supplies Record and Reflect with the decisive support for the conclusion.

Evidence is not simply collected after the question has been defined. It continually updates the auditor’s understanding, changes the relative plausibility of alternatives and may trigger additional questions or a reframing of the work.

Ask

Drives active inquiry by exposing assumptions, ambiguity, missing information, contradictions and weaknesses in explanations, methods and emerging conclusions.

Helps refine Clarify by making implicit assumptions explicit. Expands Look Beyond by uncovering additional explanations. Strengthens Evidence by testing its provenance, relevance and reliability. Improves Record and Reflect by challenging whether the reasoning can withstand scrutiny.

Asking operates throughout CLEAR rather than at a single point. Questions clarify the task, generate alternatives, interrogate evidence, test updating and challenge the final conclusion.

Record & Reflect

Makes the reasoning visible and tests it as a whole. Connects the conclusion to the purpose, alternatives, evidence, assumptions and implications, while identifying what remains uncertain or unresolved.

Gives feedback to Clarify where the conclusion does not answer the original question. Reveals missed alternatives for Look Beyond. Identifies evidence gaps or contradictions for Evidence. Generates further questions for Ask. Integrates account-area and specialist findings into the overall audit judgement.

Record and Reflect is not merely the final documentation stage. Making the reasoning explicit helps the auditor see gaps, contradictions and unsupported leaps, which may require renewed clarification, further alternatives, additional evidence or better questions.

1.2.3 Pseudo CLEAR-Thinking

Using the language without performing the thinking

There is a risk to be addressed in implementation and design: the risk that people adopt the vocabulary of CLEAR-Thinking without carrying out the mental operation the words are intended to prompt. For example, an auditor may say that they are “focused on purpose” while defining the purpose as completing the required procedure. This is pseudo-purpose: the language of purpose is present, but the work remains organised around process completion rather than a valued audit outcome.

These risks described in the table below will be highlighted along with solutions throughout the guide. 

CLEAR

Superficial imitation

What genuine thinking involves

Diagnostic test

Clarify

Restating the task or procedure as the purpose: “The purpose is to complete revenue testing.”

Identifying the audit question, the real-world claim being tested, the uncertainty to be reduced and the decision the work must support.

Does the stated purpose describe a valued audit outcome, or merely an activity to complete?

Look Beyond

Listing several generic alternatives because the template asks for them.

Identifying materially different, entity-specific explanations that could plausibly produce the observed facts and would lead to different evidence or audit responses.

Did the alternatives change what evidence was sought, what questions were asked or how the risk was assessed?

Evidence

Naming documents reviewed, describing procedures performed or stating that evidence “supports” the conclusion.

Assessing what the evidence establishes, what it does not establish, its reliability and dependencies, and whether it distinguishes between competing explanations.

Could the same evidence exist if the concerning alternative were true? Does the evidence discriminate?

Ask

Asking more questions, using challenging language or recording that management was “challenged.”

Asking questions directed at assumptions, gaps and distinctions that could change the auditor’s understanding, work or conclusion. It also includes seeking coaching, consultation or specialist support.

What answer to the question would cause the audit work or conclusion to change?

Record & Reflect

Producing a polished conclusion, completing a reflection box or answering “yes” to a challenge question.

Reconstructing the material reasoning, testing whether the conclusion follows from the evidence, identifying residual uncertainty and deciding whether closure is justified.

Did reflection expose a gap, change the work or explain why no change was necessary?

1.2.4 Leadership & Culture

CLEAR-Thinking depends on auditors being able to express uncertainty, ask questions and test emerging views before those views harden into conclusions. Yet the evidence on the “uncertainty dilemma” suggests that professionals may protect the appearance of competence even when doing so harms the quality of the outcome. At senior levels, this can be experienced as imposter syndrome. When people believe that admitting doubt could weaken their standing, they are more likely to remain silent, make untested assumptions or present confidence they do not yet possess. In audit, this can suppress the very behaviours on which sound judgement depends: surfacing gaps in understanding, considering alternative explanations, seeking support and revising a view when the evidence does not fit.

Question Would you Prefer to be seen as decisive, even if it leads to a negative outcome? Or indecisive, even if it leads to a positive outcome?

This question was put to 3,159 professionals across 25 organisations. The response is uncomfortable for anyone in a profession that trades on its authority. Two thirds of professionals, asked anonymously and without penalty, said they would rather appear decisive and accept a negative outcome than appear uncertain and reach a positive one.

Culture therefore plays a critical role in determining whether CLEAR is used as a genuine thinking discipline or merely as professional language. Leaders need to create the conditions for open dialogue through their everyday responses to uncertainty: whether they welcome questions, acknowledge the limits of their own knowledge, invite challenge before offering their view, and treat escalation as responsible judgement rather than weakness. The most powerful signals are behavioural. When senior auditors openly test assumptions, they create psychological safety, and make it safe to say, “I do not understand,” and respond constructively when concerns are raised, they reduce impression management and make uncertainty available for examination and resolution.

This is critical to audit quality because uncertainty that remains hidden cannot be clarified, tested or resolved. Open dialogue allows teams to build a more accurate understanding of the business, identify competing explanations, evaluate what evidence does and does not establish, and involve the right expertise before closure. Leadership and culture do not replace individual professional responsibility, but they shape whether auditors can exercise it effectively. CLEAR-Thinking is therefore most likely to improve audit quality where the environment rewards the quality of the reasoning and outcome, not simply the appearance of confidence/certainty/decisiveness.

High Performing Teams describes the culture of quality we are seeking to create in Audit. The Audit Specific Behaviours translate that ambition into the day-to-day expectations placed on teams, including maintaining a challenge mindset, prioritising the team and taking pride and accountability in the quality of our work.

CLEAR-Thinking provides the thinking mode that enables and strengthens those behaviours. CLEAR is therefore not a separate initiative from High Performing Teams: it helps turn the desired culture and behaviours into the way teams think, question, coach and reach judgements during the audit. Leaders embed that connection by role-modelling CLEAR-Thinking, welcoming uncertainty and challenge, and rewarding the quality of reasoning rather than merely the appearance of confidence or completion.

1.3     The role of Scaffolding

A scaffold is any feature of the working environment that helps an auditor notice the right thing and carry out the thinking a judgement requires. It may be a prompt, question, sequence, example, default, interaction or supporting information. Effective scaffolding supports the reasoning without replacing it: it makes the intended mental operation clear, keeps the conclusion open and uses the least structure needed to help the auditor think well.

1.3.1 The environment already scaffolds thinking

Thinking does not happen in isolation. What an auditor notices, questions and concludes is influenced by the language, structures and signals around the work. Test plans, templates, review notes, deadlines and performance measures all direct attention and suggest what matters.

Some of these scaffolds support good thinking, while others can narrow it too early. A step that asks the auditor to “obtain and corroborate management’s explanation” directs attention to confirming a single account. A conclusion field pre-populated with “no issues noted” makes closure the expected answer. A review note asking, “why is this still open?” may discourage the auditor from pursuing a genuine uncertainty. The environment is therefore already scaffolding thinking. The design question is whether it is doing so deliberately and in the right direction.

1.3.2 Making scaffolding deliberate

The purpose of this guide is to help designers make that scaffolding deliberate. It explains the thinking each CLEAR element is intended to promote and shows how to build that thinking into the places where audit work already happens: planning, briefing, evidence collection, coaching, review, consultation and documentation.

The aim is not to add a separate CLEAR-Thinking process. It is to work on the language, structures and interactions already shaping the work. A well-designed scaffold appears when the judgement can still be influenced, directs attention to the right issue and provides only as much support as the task and auditor require. Done well, the result should not feel like an additional compliance requirement. It should feel like the existing audit work is better framed, better questioned and easier to reason through.

1.3.3 Limitations and a caution

Language cannot compensate for missing knowledge: CLEAR-Thinking calls on business understanding and industry knowledge, along with technical accounting knowledge. Well-designed scaffolding prompts the required thinking and draws upon this knowledge; it doesn’t seek to replace the thinking or the knowledge. 

More scaffolding is not necessarily better: Excessive or poorly designed scaffolding can increase cognitive load, disrupt reasoning, encourage checklist compliance, reduce ownership, and become a source of model answers. When scaffolding becomes excessive, it’s typically because the design of templates, etc., focuses on improving quality by reducing the need for independent thinking and replacing it with tick-box procedures.

1.4     The role of Meta-cognition (Tracking Your thinking)

Meta-cognition, in the simplest terms, is "thinking about thinking." It refers to your awareness and understanding of your mental processes — and your ability to monitor, evaluate, and regulate them. In CLEAR-Thinking, it is an important practical skill for scaffolding design, coaching, and personal audit practice. 

1.4.1. In design

Designing scaffolding requires metacognition because the designer is trying to influence a cognitive process that cannot be seen directly. The task is not only to write clear language. It is to simulate, as realistically as possible, how wording, information order, defaults, examples, response formats and timing are likely to shape another person’s attention and reasoning in the circumstances of the work.

A metacognitive designer steps outside the intended message to examine its likely mental effect. They ask: What does this wording make psychologically salient? Which mental operation will it prompt? What answer does the structure make easiest? Could the prompt be completed professionally without the intended thinking occurring?

This simulation also requires the designer to examine their own thinking. Metacognition helps the designer distinguish the effect they hope the intervention will have from the effect its actual language and structure are likely to produce.

 1.4.2. In coaching (and learning design)

Coaching depends on two related metacognitive abilities. The first is access to one’s own reasoning. Experienced auditors often recognise a problem rapidly, but their knowledge has become compressed: they see that something does not fit, yet cannot immediately explain the sequence of observations, expectations and comparisons that produced the judgement. Effective coaching requires the coach to unpack that expertise and make it available to another person.

This means the coach can externalise the how and what of their thinking: what they noticed, what real-world picture they constructed, how they constructed it, which prior knowledge they drew on, what alternatives developed, how they developed them, which they kept open and why, how did they know to close them, what evidence they expected, how did they build those expectations, where confidence changed and what would cause them to revise the view. Sharing or coaching this thinking gives the learner access to the cognitive moves behind the conclusion rather than only the conclusion itself.

The second ability is to develop a working model of how the other person may be thinking. The coach infers the learner’s current representation from what they say, what they omit, what they ask, and how they respond to carefully crafted questions. This cognitive modelling matters because it enables the coach to provide the precise support needed: a question, a comparison, a demonstration, contextual knowledge, technical input, or permission to remain uncertain.

What is lost without it: the coach supplies answers, gives review notes aimed only at the output, or asks broad questions such as “do you understand?”. The learner may reproduce the senior’s wording without acquiring the reasoning, while the coach mistakes agreement or a polished file for genuine understanding.

1.4.3 In audit practice

For an individual auditor, metacognition means tracking how their thinking is developing while the work is still capable of changing. The auditor monitors not only what they know, but how they are arriving at the view: the purpose currently guiding the work, the mental picture of the business, the explanations being held open, the expectations used to interpret evidence, the source and calibration of confidence, and the points that remain difficult to explain.

Useful metacognitive signals include surprise, confusion, unusual ease, unjustified comfort, discomfort that cannot yet be articulated, reliance on a fluent management narrative, and a growing volume of evidence without a clearer judgement. These signals indicate where attention or regulation may be needed.

Part 2

Designing for CLEAR-Thinking

2.1 Experience-Led Design (ELD)

2.1.1 Introduction

Part 1 described the thinking that CLEAR is intended to strengthen. Part 2 looks at how designers can build that thinking into the flow of work, so that the right thinking is more likely to happen at the time it matters most.

The approach used in this guide is Experience-Led Design (ELD).

Experience-Led Design starts with the auditor's experience rather than with the material, tool or system being designed. Instead of beginning by asking How should we improve this template? What should this guidance say? or What question should we add?, the designer first considers what is happening from the auditor's perspective: where they are in the work, what they are trying to achieve, what they are attending to, what they are likely to be thinking, and how the surrounding experience may be shaping that thinking.

The purpose is not just to make CLEAR more visible in audit materials. It is to make CLEAR Thinking more likely.

This distinction matters. Adding a question that says, “Have you applied professional scepticism?”, “Have you Looked Beyond?” makes the desired behaviour visible, but does little to activate it. A question such as “What evidence would distinguish management's explanation from the strongest credible alternative?” provides a much more specific scaffold for the attention and mental operation required. Language, structure, sequence and interaction can therefore influence thinking without explicitly referring to CLEAR at all.

Experience-Led Design combines three core skills applied throughout the process that follows:

Taking perspectives — experiencing the work from different positions. The designer deliberately steps out of their own expert perspective and into that of the auditor who will encounter the design. This helps reveal assumptions about knowledge, context and intent that may otherwise remain invisible.

Metacognition — noticing and using thinking as design evidence. The designer pays attention to their own cognitive response as they experience an instruction, interaction or task. What did their attention go to? What did they assume? What question appeared internally? Where did their thinking narrow or become uncertain? This makes the designer's own experience a testing lab for understanding and refining the design.

Language as scaffolding — deliberately shaping attention and mental operations. Language does more than communicate information. Wording, structure and sequence influence what becomes salient, how a problem is represented and what mental operation someone performs next. Effective scaffolding therefore specifies and supports the thinking required rather than merely naming desirable qualities such as “challenge”, “consider” or “apply scepticism”. (See activating thinking through language section in the appendix)

2.1.2 The Experience-Led Design process

The ELD process is organised around three connected stages:

DIAGNOSE → DEFINE → DESIGN

Critical intervention point → Desired thinking → Designed experience

Diagnose prevents the designer from starting with a solution before understanding the auditor's experience. Define prevents vague ambitions such as “improve clarity” from becoming the design objective. Design then focuses effort on finding the most effective way of activating the required thinking through the available touchpoints.

The process is deliberately iterative rather than a rigid sequence. Designing or testing an intervention may reveal that the original diagnosis was incomplete or that the thinking objective needs to change. The designer should move backwards when new insight warrants it.

This is consistent with CLEAR Thinking itself. CLEAR is not intended to operate as five boxes completed once in sequence. Later evidence, inquiry or reflection can legitimately cause the auditor to revisit earlier framing or explanations.

2.2 Diagnose — Identify critical intervention points

WHEN / WHERE could scaffolding CLEAR Thinking make a difference?

The first phase of ELD is to identify the points in the auditor's experience where design intervention could make a meaningful difference to thinking.

This is broader than looking for things that are currently “wrong”. A critical intervention point may arise because something in the existing experience is actively inhibiting CLEAR Thinking. It may also arise because the auditor reaches an important judgement with relatively little support, creating an opportunity for appropriately designed scaffolding to strengthen the thinking.

The designer is therefore looking for points of leverage: moments where the nature of the experience has the potential to materially affect the thinking that follows.

2.2.1 Experience the work from the auditor's perspective

Start with the experience, not the artefact.

A designer responsible for an APT file, standardised workbook, guidance document or learning intervention will naturally see the work through that design surface. The auditor does not. They experience a flow of work in which tools, information, conversations, deadlines, team interactions, reviewer expectations and prior knowledge combine.

The designer should therefore mentally step into that flow and ask:

What is happening to the auditor at this point? What are they trying to achieve? What do they know? What have they just seen or been told? What are they likely to attend to next? What assumptions might they make?

The level of specificity matters.

“Planning” is generally too broad to be a useful intervention point. “When work is being assigned to a team member” is much more useful. Likewise, “impairment testing” describes an area of work. “When the auditor first encounters management's forecast and explanation for growth” describes a moment at which the auditor's subsequent reasoning may be shaped.

Thinking at this level helps the designer understand when and where the experience has leverage over cognition. It also avoids the risk of focusing on general improvements – e.g. updating the standard templates against CLEAR. These will be updated, but as a product of designing for the moments an auditor engages with them.

2.2.2 Identify opportunities to support CLEAR Thinking

As the designer moves through an auditor’s journey through an audit, they should look for moments where CLEAR Thinking could benefit from scaffolding.

Sometimes the opportunity will be visible because the existing experience is pulling thinking in an unhelpful direction. For example, an instruction to “obtain support for management's explanation of the revenue increase” makes management's explanation the organising frame before the auditor has considered whether other explanations are plausible. Language of this kind can scaffold justification and confirmation rather than open enquiry.

In other cases, there may be no obvious design failure. The significance of the moment itself may justify additional support. An auditor may be moving from understanding an estimate to deciding what evidence to obtain, for example, and a relatively light scaffold at that point could materially improve the quality of the subsequent work.

Useful signals include moments where the auditor is likely to:

• lose sight of the purpose and default to completing a procedure;

• remain within accounting abstractions rather than connect numbers to the underlying business reality;

• become anchored to the first available explanation;

• move into evidence gathering before establishing what the evidence needs to distinguish;

• accept information without noticing an important contradiction or limitation;

• need to recognise the limits of their own knowledge and seek support;

• reach closure without sufficiently updating or testing the overall reasoning.

These are not a checklist of mandatory intervention points. They are clues to where the experience may deserve closer attention.

Connected thinking is particularly important in identifying such moments. Financial information is an abstract representation of business activity. Auditors need to be able to move from numbers and accounting treatments to the customers, transactions, events, decisions, assets and processes that produced them, and from that reality form expectations against which evidence can be assessed.

2.2.3 Prioritise critical intervention points

Not every opportunity should move into design.

Attempting to scaffold every instance of judgement would create unnecessary complexity and could itself inhibit thinking. Excessive scaffolding can increase cognitive load, encourage checklist completion and reduce ownership of the reasoning. The objective is the least intrusive scaffold capable of improving the relevant thinking.

The designer should therefore consider where intervention is likely to have sufficient value. Relevant considerations include:

• the significance of the judgement or outcome;

• how vulnerable the thinking is at that point;

• whether what happens at the moment affects substantial work downstream;

• how frequently the moment occurs and how many auditors encounter it;

• the strength of the opportunity to influence thinking;

• whether another part of the experience already provides sufficient support;

• the likely burden created by intervention.

The output from Diagnose should be a prioritised set of critical intervention points, described with enough specificity that the designer can understand what the auditor is experiencing and why the moment matters.

2.3 Define — Specify the thinking required

WHAT CLEAR Thinking is needed here?

Having identified a critical intervention point, the next task is to define what needs to happen in the auditor's thinking at that moment. This step is essential because broad design objectives such as “encourage professional scepticism”, “improve judgement” or even “support Look Beyond” are not sufficiently precise to design against. The designer needs to move from the desired quality of thinking to the cognitive work that would produce it.

2.3.1 Use CLEAR to help identify the required cognitive move(s)

CLEAR provides the primary framework for defining the thinking objective. The designer should ask which element, or combination of elements, is most relevant to the intervention point:

Clarify where the auditor needs to establish the real question, purpose, uncertainty or underlying business claim.

Look beyond where the auditor needs to move beyond an initial explanation and keep materially plausible alternatives available.

Evidence where the auditor needs to determine what evidence would discriminate between possibilities, understand what evidence establishes and recognise what remains unresolved.

Ask where inquiry, self-questioning, coaching, consultation or specialist input is needed to change understanding or resolve a material gap.

Record & reflect where the auditor needs to update the reasoning, test whether the conclusion follows, identify residual uncertainty or make the material reasoning visible.

The aim is not to find a reason to insert all five CLEAR elements. The strongest design may support one highly specific cognitive move. Where several CLEAR moves genuinely need to operate together, they should form a connected reasoning sequence in which one move changes the conditions for the next.

2.3.2 Define the target thinking

Begin by stepping into the desired experience; rather than asking only “What should the auditor do?”, ask:

What would I need to notice here? What would I need to understand? What question would I be asking myself? What would I need to compare, reconstruct, distinguish, predict, test or update?

This is an important use of metacognition in ELD. The designer deliberately enters the form of thinking they are trying to support and observes the internal moves that keep that thinking on track. For example, “apply professional scepticism to management's explanation” names a desirable quality but leaves the required mental operation unspecified. A more precise thinking objective might be:

Separate the observed revenue movement from management's explanation, construct the strongest materially different explanation that could also produce the pattern, and identify what evidence would distinguish between them.

That provides something concrete for the design to activate and something observable against which it can later be tested.

2.3.3 Keep the thinking connected to reality

As the thinking objective becomes more precise, the designer should test whether it remains connected to the underlying business reality.

Connected thinking is the ability to connect financial information, accounting treatments and audit evidence to the real business events, activities, people, decisions, assets and processes that produced them. It enables the auditor to ask not only whether something is technically correct but whether it makes sense in the specific circumstances of the business.

This matters across CLEAR. A Clarify intervention may need to help the auditor understand what a balance is actually claiming about the business. Look Beyond may require alternative business realities rather than generic audit risks. Evidence may require the auditor to predict what would be observable in the business if each explanation were true.

The output from Define should therefore be a specific thinking objective: a description of what the auditor needs to notice and do mentally at the intervention point, sufficiently clear that the designer can design for it and later test whether it occurred.

2.4 Design — Influence and activate the thinking

HOW can we enable the required thinking?

Once the intervention points and thinking objectives are clear, attention moves to the experience itself. Design asks how the surrounding experience at a particular moment can make the desired thinking more likely. This is where the designer moves from diagnosing and specifying cognition to shaping the environment in which that cognition takes place.

2.4.1 Identify and consider the effect of other relevant touchpoints

The same thinking moment can be influenced through different parts of the auditor's experience. Potential touchpoints include:

• tools, templates and audit artefacts;

• systems and workflow;

• the information presented to the auditor and how it is structured;

• team interactions and coaching;

• review;

• leadership behaviour;

• learning events and materials;

• guidance and examples.

The designer may only have direct responsibility for one of these. Someone redesigning an APT file, for example, may not control the training programme or the way a manager coaches the task. Experience-Led Design nevertheless requires them to understand that their design sits within this wider experience. This is why design needs to be a team activity with effective coordination by an overall design architect.

2.4.2 Five-step detailed design process

Detailed design of an intervention follows the five-step process shown below and described in the following subsections.  

2.4.3 Design target

The form of thinking being designed for was identified during the Define stage (see section 2.3.2 above). It is revisited here because the person undertaking the detailed design may not have been involved in the earlier Diagnose and Define stages. Before designing the intervention, the designer therefore needs to step into the relevant audit moment and experience the target form of thinking for themselves. This creates a clear cognitive reference point for the design: what the auditor needs to notice, question, connect, test or reconsider at that moment, and what successful support for that thinking should feel like in practice.

2.4.4 Evaluate the current experience

Before creating a solution, experience the current design.

The designer should encounter the instruction, template, interaction or material as closely as possible to the way the auditor encounters it and notice what happens to their own thinking. This is where the three ELD skills begin to work together.

·       Use perceptual positions to move from the designer's intent into the auditor's experience.

·       Use metacognition to notice what the current design does: where attention goes, what assumptions arise, what feels already decided, what questions occur internally and where thinking becomes difficult or closes prematurely.

·       Use understanding of language as scaffolding to examine how wording, structure and sequence may be producing those effects.

The question is not simply whether the current design is clear or well written. It is:

What thinking does this design actually make more likely?

This can expose gaps between intention and effect. A current design may intend an instruction to encourage scepticism while discovering that its wording anchors the auditor to management's explanation. A template intended to encourage reflection may position the reflective question only after the conclusion has been reached, making retrospective justification more likely than live reflection.

2.4.5 Create and refine the intervention

The designer can now begin to change the experience using their own experience as a design tool.

The aim is to find the minimum effective scaffold that makes the defined thinking more likely while preserving the auditor's responsibility for performing the reasoning.

That scaffold may be a question, label or prompt, but it need not be. It could involve changing the sequence in which information is presented, separating observation from explanation, introducing a comparison, altering the structure of a template, changing a default, providing a worked contrast, creating a visual representation, changing a coaching interaction or introducing a prompt at a particular point in a digital workflow.

Language deserves particular attention because relatively small changes can alter the cognitive task. Broad verbs such as consider, understand and challenge leave the required operation largely unspecified. More operational wording — reconstruct, compare, distinguish, predict, trace — can provide a clearer internal instruction.

Refinement should be experiential rather than purely editorial. The designer repeatedly steps into the intervention and asks:

What happens in my thinking now? What does this make me notice first? What am I assuming? What am I saying to myself internally to perform the desired thinking? Does the design evoke that process, or merely describe it? What is missing? What is unnecessary?

This creates a rapid design cycle:

experience → notice your thinking → change the scaffold → experience it again → notice what changed.

The designer’s own cognition becomes a provisional testing environment in which early refinements can be explored cheaply and repeatedly before being put in front of other people. Metacognition is especially useful here because seemingly small differences in wording can cause quite different internal responses.

The insight generated through metacognition may indicate that the problem cannot be solved by wording alone. Perhaps the designer discovers that they can only answer the prompt after looking elsewhere for information about the entity. The refinement may therefore need to place relevant operational information alongside the question.

It should also expose friction: anything in the experience that makes the desired thinking unnecessarily difficult. Friction may come from excessive navigation, duplicated information, unclear language, poorly timed prompts, unnecessary documentation, competing instructions or simply too much scaffolding.

Removing friction does not mean making the work cognitively effortless. Productive audit thinking can be demanding. The aim is to remove unnecessary difficulty around the thinking, not the thinking itself.

Preserve the auditor’s thinking: There is an important boundary. The purpose of the designer accessing their own expert cognition is to identify how to activate the auditor’s thinking, not to insert their thinking into the material for them. A designer may realise that they themselves would consider three particular explanations. It does not automatically follow that those explanations should be provided as a drop-down list. Doing so may replace the very generation and comparison process the scaffold is intended to support.

2.4.6 Test the thinking produced

The purpose of the intervention is to influence thinking; testing should therefore return directly to the thinking objective established earlier in the design process and examine whether the revised intervention is producing that thinking when it is encountered in context.

The core design question is:

When an intended auditor experiences this revised design in the relevant work moment, what happens in their thinking, and is this the thinking we intended to activate?

Translate the target thinking into observable signs

Before testing the intervention, the designer should return to the target form of thinking and consider what they would expect to observe if that thinking were taking place. The mental operation itself cannot be seen directly, so the designer needs to identify the language, questions, actions, decisions and changes in direction that would provide reasonable evidence that it is occurring.

For example, if the intended thinking is to keep alternative explanations open for longer, the designer might expect the auditor to generate more than one plausible explanation, distinguish between them and seek evidence that could discriminate between them. If the intended thinking is more connected, the designer might expect the auditor to move from an abstract number or accounting treatment into the underlying business events, people, processes or constraints and begin forming expectations about what should be observable in practice.

The purpose is not to create another checklist of behaviours. It is to establish, before testing, what evidence would indicate that the intervention is influencing the underlying thinking rather than merely changing the form of the response.

Use the sounding board as an early metacognitive test

The sounding board provides an important bridge between the designer’s own refinement process and testing with intended users. Because sounding-board members share an understanding of CLEAR-Thinking and the design intent, they can experience the intervention themselves and pay deliberate attention to what it does to their thinking.

Rather than simply commenting on whether wording is clear or whether a template looks usable, they should notice their own cognitive response. What did the intervention cause them to attend to? What question appeared in their mind? Did it make them construct a clearer picture of the underlying situation? Did it open up alternative explanations, direct them towards particular evidence, or cause them to reconsider an initial conclusion? Equally importantly, did anything in the design pull them towards premature closure, procedural completion or another unintended form of thinking?

This metacognitive feedback can help the designer refine both the intervention and the observable signs that will subsequently be looked for when testing with intended users.

Test across different intended users

The effectiveness of a scaffold will depend partly on the knowledge and experience of the person encountering it. A short question may activate a rich chain of reasoning for an experienced manager because they can supply the relevant audit, business and causal knowledge themselves. The same question may leave a less experienced auditor unsure what they are being asked to think about or without sufficient knowledge to perform the required mental operation.

The opposite can also occur. A detailed scaffold that helps an inexperienced auditor construct the problem may interrupt, slow or unnecessarily constrain the reasoning of someone more experienced. Testing should therefore include the different user groups for whom the intervention is intended, rather than assuming that the same design will affect everyone in the same way.

Use Test to learn, not simply to approve

Testing should not be treated as the final approval stage before implementation. It remains part of the iterative design process. Its purpose is to generate evidence about the relationship between the design and the thinking it produces. Where testing reveals an unintended effect, return to the experience. The problem may lie in the intervention, but it may also reveal that the original thinking objective or diagnosis was incomplete.

2.4.7 Embed, learn and adapt

Once an intervention performs well enough to take forward, it needs to work within the reality of audit work. A design that produces the intended thinking during Test may behave differently when it is encountered alongside deadlines, prior-year files, system constraints, manager instructions, review expectations and established ways of working.

Embedding therefore involves more than publishing a revised template, adding a question or changing an instruction. Where practical, the intervention should first be introduced within a bounded part of the real working environment so that the designer can observe what happens when auditors encounter it naturally, in the moment for which it was designed.

The focus remains the thinking produced. The designer should ask:

What happens to the intended thinking when this intervention becomes part of the auditor’s real working environment, and what do we need to learn and change as a result?

Consider the whole experience around the intervention: An intervention does not operate independently of the other touchpoints surrounding it. A template may encourage auditors to consider alternative explanations, while the work-allocation conversation presents management’s explanation as the accepted starting point. Guidance may encourage uncertainty to be surfaced, while review behaviour signals that unresolved matters should be closed quickly. A prompt may encourage connected thinking, while the system makes the information needed to build that connection difficult to access.

The designer should therefore consider whether the surrounding touchpoints reinforce or compete with the intended thinking. Where another touchpoint is exerting the stronger influence, adding further prompts or instructions may not solve the problem. The wider experience may need to change.

The objective is not identical CLEAR wording across every touchpoint. It is a coherent experience in which systems, templates, communications, learning, coaching, team interaction and review collectively support compatible forms of thinking.

Learn from what users actually do: Real use will often expose effects that were difficult to reproduce during Test. Auditors may overlook the intervention because other information is competing for attention. They may understand the question but continue rolling forward prior-year wording because it is easier. They may identify uncertainty but avoid asking for help because the working environment discourages interruption or escalation.

These behaviours are useful design information. They should not automatically be interpreted as resistance or failure to follow the new process. Instead, the designer should ask what the behaviour reveals about the experience that has actually been created. A workaround may reveal friction. Repeated copying may indicate that the response format rewards completion rather than thought. A useful question that is ignored may be appearing at the wrong moment or through the wrong touchpoint.

Look for evidence that thinking and behaviour have changed: Measures of successful embedding should go beyond whether the intervention has been seen, used or completed. High completion may coexist with little or no change in thinking.

The designer should return to the target thinking and look for practical evidence that the desired cognitive shift is appearing in real work. Depending on the intervention, this may be visible in the questions auditors ask, the connections they make to business reality, the alternative explanations they consider, the evidence they seek, how they respond when evidence does not fit, when they seek support, or how their conclusions change as their understanding develops.

The underlying principle is:

Measure evidence that the thinking and resulting action changed, not simply evidence that the scaffold was completed.

Continue to learn and adapt: Embedding is not the end of the design process. The audit environment changes, users adapt to interventions, and apparently effective designs can gradually become routine completion mechanisms. The designer should therefore continue to step into the user experience and examine what the intervention is doing to thinking over time.

Where the intended thinking is no longer occurring, the breakdown should be treated as information about the design and the wider system. Is the scaffold now too weak, too strong or badly timed? Has another touchpoint become more influential? Has familiarity changed how it is interpreted? Is unnecessary friction encouraging a workaround? Is the knowledge required to perform the thinking actually available?

The answer determines where the design process needs to return. Evidence from real use should feed back into Diagnose, Define and Design, allowing both the intervention and the wider experience around it to continue to improve.

Embed therefore does not simply put a new question, template or process into practice. Its purpose is to help the desired way of thinking become part of the experience of doing the audit — while continuing to learn whether the design is genuinely supporting that thinking as the environment and its users change.

Part 3

CLEAR Thinking Element Descriptions

Design Reference

to be

updated based

on learning from application

3.1 The foundations – connected thinking: linking numbers to reality

Connected thinking helps auditors move between financial representations and the business events, activities, people, decisions, assets and processes that produced them.

The two-way movement

Direction

Designer should help the auditor ask

Business to numbers

What happened in the business, and how should that appear in the accounting records?

Numbers to business

What events, activities and assumptions would have to exist for this number to be reasonable?

Connected thinking as a practical cycle

1.     Notice the relevant facts: who, what, when, where, decisions, changes, resources and constraints.

2.     Connect the observations into sequences, dependencies and cause-and-effect relationships.

3.     Build a provisional business narrative that explains how the number or estimate arose.

4.     Form expectations about what should be visible if the account is true.

5.     Compare those expectations with reliable evidence.

6.     Update the account, evidence strategy or conclusion when observations do not fit.

Design implications

1.         Use concrete referents: customers, products, contracts, dates, sites, decisions, goods, services and cash.

2.         Separate observed facts from management explanation, auditor inference and prior experience.

3.         Ask for the operating driver behind multiple assumptions rather than reviewing each input in isolation.

4.         Use walkthroughs, operational observation and specialist access when language alone cannot supply the needed knowledge.

5.         Require testing of a plausible story; do not reward narrative fluency by itself.

3.2 Clarify

Clarify establishes the audit question, why it matters and the real-world claim behind the financial representation. It protects the work from drifting into procedure completion.

Design dimension

Controlled guidance

Practical purpose

Create a usable task model: what must be understood or concluded, what uncertainty matters, what real condition is represented, and what would count as a satisfactory answer.

Shift in attention

From the assigned step or document to the assurance purpose, underlying claim, decision and limits of the planned work.

Mental operations

Define, bound, translate, reconstruct, specify success criteria and identify what the planned procedure cannot establish.

How connected thinking contributes

Clarification becomes concrete when the auditor can state what the number claims about real customers, transactions, assets, decisions, events or future activity.

Productive language patterns

·    What exactly must this work establish, and why does it matter to the audit?

·    What claim about business events/activities/conditions etc sits behind this number or treatment?

·    What uncertainty must be reduced before a conclusion is possible?

·    What will the planned procedure establish, and what will it not establish?

·    What result would require the task to be reframed or escalated?

Coaching and review behaviours

·    Ask the auditor to explain the task in their own words before supplying the answer.

·    Test whether they can distinguish the audit purpose from the assigned procedure.

·    Ask what the balance or transaction claims about the business.

·    Notice when the person describes only the file, calculation or methodology.

·    Reframe collaboratively when the task has changed rather than treating revision as an error.

Weak or misleading patterns

Pattern

Why it weakens thinking

"Confirm the balance is reasonable."

The conclusion is presupposed and the real claim is not specified.

A long background section followed by a generic objective

Information volume does not create a usable question.

Clarify field completed only at the end

The wording can rationalise the work but cannot guide it.

What good looks like

·    The auditor can state the question and why it matters without repeating the procedure.

·    The financial number is expressed as a claim about real activity or future events.

·    The work that follows is streamlined by answering the question at hand, and not completing a generic set of unnecessary procedures.

·    The limits of the planned work are understood.

·    The auditor notices and corrects drift from purpose to completion.

·    A changed fact can trigger a revised question or plan.

Common superficial imitations

·    Restating the template objective in different words.

·    Describing the procedure as the purpose.

·    Using generic risk language that could apply to any entity.

·    Producing an elaborate business narrative without a bounded audit question.

Connections to other CLEAR moves

Clarify creates the object that Look Beyond explains and Evidence must resolve. Evidence may reveal that the original question was too broad, too narrow or directed at the wrong uncertainty. Record and Reflect tests whether the final work still answers the clarified question.

Design contrasts

Weak design

Near miss

Stronger design

Why the difference matters

"Test revenue cut-off."

"Understand year-end revenue."

"For sales recorded in the final ten days, determine whether control transferred before year end and identify any pattern that indicates wider timing risk."

The stronger version defines population, real event, decision and wider consequence.

"Review the impairment model."

"Assess whether assumptions are reasonable."

"Determine whether the operating conditions needed to generate the forecast cash flows are supported and whether a plausible change would remove headroom."

It shifts from file review to the business condition and decision threshold.

2.4 Look Beyond

Look Beyond preserves appropriate openness by separating observation from explanation and identifying materially different plausible realities before one account organises all later evidence.

Design dimension

Controlled guidance

Practical purpose

Develop a bounded set of entity-specific accounts that could explain the observation and would lead to different evidence needs, risks or audit responses.

Shift in attention

From the first or most fluent explanation to the distinction between what is observed and what could plausibly produce it.

Mental operations

Separate, generate, compare, rank, prune, counterfactually simulate and identify the strongest credible alternative.

How connected thinking contributes

Realistic alternatives come from the entity's customers, products, contracts, capacity, incentives, processes, decisions and cash flows rather than generic fraud or error lists.

Productive language patterns

·    What materially different business realities could produce this pattern?

·    Could the number be correct while the explanation is wrong?

·    What is the strongest credible alternative, not merely another possibility?

·    Which explanation is most likely, and which is most concerning?

·    Which alternative would change the audit response if true?

·    What are we treating as observation that is actually interpretation?

Productive structural patterns

·    Present observations separately from management's explanation where anchoring risk is material.

·    Use two or three comparative columns only when accounts predict different evidence.

·    Require a short statement of the audit consequence for each material account.

·    Allow alternatives to be combined, revised or removed as evidence develops.

·    Use examples that demonstrate specificity without supplying reusable alternatives.

Coaching and review behaviours

·    Ask for the strongest alternative before offering your own.

·    Probe whether each account is entity-specific and capable of changing the evidence plan.

·    Distinguish unlikely from implausible; a concerning account may deserve attention despite lower probability.

·    Stop the exercise when the field is bounded; do not reward length.

·    Reopen the field when surprising evidence cannot be explained by the current accounts

Weak or misleading patterns

Pattern

Why it weakens thinking

"List five alternative explanations."

A numerical target rewards quantity and token possibilities.

"Challenge management's explanation."

The management account remains the organising frame and the required operation is unspecified.

Generic fraud-risk menus

Pre-populated labels can replace entity-specific causal reasoning.

Alternative field placed after evidence and conclusion

The account is reconstructed retrospectively rather than influencing selection.

What good looks like

·    The auditor can articulate a small number of materially different accounts.

·    Each account is linked to the entity and the observed pattern.

·    The accounts produce different expected traces or audit consequences.

·    The auditor can say which is likely, which is concerning and why.

·    New evidence can add, weaken, merge or remove an account.

Common superficial imitations

·    Listing generic error, fraud and timing possibilities.

·    Inventing implausible alternatives to satisfy a field.

·    Using different wording for the same underlying account.

·    Recording alternatives that do not change evidence, inquiry or action.

Connections to other CLEAR moves

Clarify defines the uncertainty that Look Beyond explains. The accounts should hand directly to Evidence by identifying where observable traces differ. Ask probes the unresolved distinction. Record and Reflect shows how concerning alternatives were resolved or why residual uncertainty is acceptable.

Design contrasts

Weak design

Near miss

Stronger design

Why the difference matters

"Explain the 18% revenue increase."

"What else could explain the increase?"

"Revenue rose 18%, concentrated in the final six weeks, while receivables rose faster and margin fell. Which materially different business events could produce the combined pattern, and which would change our audit response?"

The stronger version preserves the observation, integrates the pattern and requires consequence-bearing accounts.

"Consider an alternative forecast."

"What if growth is lower?"

"Which operating driver supports the growth and margin assumptions together, and what credible failure of that driver would remove headroom?"

It creates a causal alternative rather than a mechanical sensitivity only.

2.5 Evidence

Evidence is designed as a means of reducing material uncertainty and distinguishing between live accounts, not as a collection of documents consistent with the preferred answer.

Design dimension

Controlled guidance

Practical purpose

Help the auditor identify what a source establishes, what it cannot establish, which accounts it is compatible with, how reliable and independent it is, and what it changes.

Shift in attention

From accumulating support to evaluating evidence credibility, evidential contribution, limitation, provenance, dependency and diagnostic value.

Mental operations

Predict, trace, compare, discriminate, reconcile, evaluate reliability, identify dependency and update confidence or action.

How connected thinking contributes

Connected thinking links evidence to expected business traces: customer behaviour, dispatch, capacity, service delivery, disputes, cash, operational decisions and later outcomes.

Productive language patterns

·    What does this evidence establish, and what does it leave unresolved?

·    Which live explanations are compatible with this evidence?

·    What does this source help distinguish?

·    Could the same evidence exist under both accounts?

·    What management input or system dependency sits behind this source?

·    Is this IPE (Information Produced by the Entity)?

·    What did this evidence change in our understanding, plan or conclusion?

·    What observation did none of the current accounts predict?

Coaching and review behaviours

·    Ask what the source cannot establish before asking whether it is sufficient.

·    Probe why apparently independent evidence may rest on the same input.

·    Treat surprising evidence as information about the model, not an obstacle to clear away.

·    Ask which next source would be most diagnostic rather than requesting more evidence generally.

·    Escalate when the evidence remains weak or the specialist judgement exceeds the team's competence.

Weak or misleading patterns

Pattern

Why it weakens thinking

"Obtain support for management's explanation."

Search is restricted to confirming material.

Evidence checklist with no purpose link

Procedure output is accepted irrespective of the uncertainty.

"No exceptions noted" as the main conclusion

Absence of recorded exceptions is mistaken for resolution.

Reliability assessed after evidence selection

Evaluation can become a retrospective defence of chosen sources.

What good looks like

·    The auditor explains the contribution and limitation of decisive evidence.

·    Evidence is selected because it addresses a material distinction.

·    Dependencies and management influence are recognised.

·    Unexpected evidence changes the account, plan, confidence or conclusion where appropriate.

·    The team knows when additional evidence is unlikely to add value.

Common superficial imitations

·    Listing evidence without explaining its relevance.

·    Calling a source independent because it is in a different document.

·    Explaining away mismatches without updating.

·    Treating subsequent cash as proof of every assertion relating to the original transaction.

Connections to other CLEAR moves

Look Beyond should produce the accounts that Evidence needs to distinguish. Ask obtains missing facts and probes source limitations. Evidence may require Clarify to change the audit question. Record and Reflect identifies decisive evidence, residual gaps and why closure follows.

Design contrasts

Weak design

Near miss

Stronger design

Why the difference matters

"Agree receivables to subsequent cash."

"Document whether cash was received."

"For each selected balance, state what the post-year-end event establishes, what it does not establish, and whether the customer behaviour indicates a wider recoverability, dispute, credit-note or revenue issue."

It makes the evidential function and wider consequence explicit.

"Inspect board-approved budgets."

"Assess budget reliability."

"Identify the operating drivers that support the forecast, compare them with actual capacity and recent performance, and explain which material uncertainty remains after the comparison."

Approval and internal consistency are not treated as external evidence of achievability.

2.6 Ask

Ask uses questions, coaching, consultation and expert support to expose assumptions, obtain missing facts, improve the mental model and resolve the material distinction.

Design dimension

Controlled guidance

Practical purpose

Direct inquiry towards information or perspective that can change understanding, evidence strategy, confidence, action or conclusion.

Shift in attention

From asking for more support for the current account to asking the question, person or source most capable of resolving the gap.

Mental operations

Probe, expose assumptions, seek context, test reasoning, self-question, consult, escalate and invite alternative perspectives.

How connected thinking contributes

Connected questions seek operational knowledge from people who know how transactions, capacity, customers, systems or decisions work, not only from those who own the accounting explanation.

Productive language patterns

·    What are we treating as true without having established it?

·    Which factual point would most change the assessment?

·    What question would distinguish the two remaining accounts?

·    Who has direct knowledge of the business event or process (not necessarily someone in the finance team, someone close to the operations involved instead)?

·    What are we least able to explain?

·    Is our comfort coming from evidence, authority, familiarity or a tidy file?

·    Do we need coaching, technical support, consultation or specialist involvement?

Coaching and review behaviours

·    Ask the auditor what they think before supplying the senior view.

·    Use follow-up questions that test the causal or evidential connection.

·    Model willingness to say "I do not yet understand" and treat it as a trigger for action.

·    Invite challenge from juniors and specialists or experts where they hold relevant knowledge.

·    Persist constructively when the answer does not address the distinction; do not equate politeness with weak challenge.

Weak or misleading patterns

Pattern

Why it weakens thinking

"Challenge management."

The required question and distinction remain unspecified.

Repeated request for documents supporting the same story

Inquiry intensifies confirmation rather than broadening understanding.

Escalation only after the file is complete

Support arrives too late to shape the work.

Senior gives the correct answer immediately

The immediate task may be solved but the team's reasoning remains invisible and undeveloped.

What good looks like

·    Questions are linked to a material assumption, account difference or evidence gap.

·    The auditor can explain why the answer might change the work.

·    The team seeks operational and specialist knowledge at the right time.

·    Comfort and confidence are examined rather than asserted.

·    Answers and conversations update the model or are recognised as non-diagnostic.

Common superficial imitations

·    Recording that management was challenged without the substance or consequence.

·    Asking open questions that invite a fluent story but do not test it.

·    Escalating responsibility rather than seeking support.

·    Using "what else?" after the preferred explanation has already closed the evidence search.

Connections to other CLEAR moves

Evidence mismatch often generates the most useful Ask prompt. Ask may reveal a new Look Beyond account or expose that Clarify was incomplete. Record and Reflect should preserve only the material question, answer and consequence needed to understand the judgement.

Design contrasts

Weak design

Near miss

Stronger design

Why the difference matters

"Ask management why margin will improve."

"Challenge the margin assumption."

"Which specific price, volume, mix or cost changes produce the margin improvement; who owns each change; and where should early evidence of delivery already be visible?"

The stronger question decomposes the claim and predicts observable traces.

"Seek specialist input if needed."

"Consider whether consultation is required."

"If the conclusion depends on enforceability, valuation or technical behaviour outside the team's competence, pause and route the defined question and current evidence to the relevant specialist before closure."

It creates a usable trigger and a bounded consultation question.

2.7 Record and Reflect

Record and Reflect makes material reasoning reconstructable and uses explanation as a live test of whether the conclusion follows from the decisive evidence and whether closure is justified.

Design dimension

Controlled guidance

Practical purpose

Reconstruct the reasoning, identify decisive evidence and residual uncertainty, test coherence, regulate confidence and decide whether to close, recur or escalate.

Shift in attention

From documenting completed activity to testing the relationship between question, plausible accounts, evidence, updating and conclusion.

Mental operations

Reconstruct, explain, integrate, calibrate, identify the weakest link, test closure, anticipate challenge and revisit earlier moves.

How connected thinking contributes

A narratable conclusion explains how the financial result relates to real events and why the evidence supports that account rather than merely describing procedures and calculations.

Productive language patterns

·    What was the question and why did it matter?

·    What evidence was decisive rather than merely present?

·    What changed during the work?

·    What credible alternative remained, and how was it resolved?

·    What is the weakest link in the conclusion?

·    What fact would change the conclusion?

·    Does the conclusion follow from the evidence, or merely fit it?

·    Is closure defensible and proportionate, or simply convenient?

Coaching and review behaviours

·    Ask the auditor to narrate the reasoning before editing the wording.

·    Test the weakest link and the stopping reason rather than only grammar and completeness.

·    Distinguish a missing documentation link from a missing thinking or evidence link.

·    Do not rewrite the conclusion in a way that hides the original uncertainty or team learning.

·    Recognise justified confidence and stop unnecessary work when the purpose is met.

Weak or misleading patterns

Pattern

Why it weakens thinking

"Summarise procedures performed and conclude."

Activity becomes the narrative and the judgement link remains hidden.

"Document how professional scepticism was applied."

The file may reward performative challenge after the event.

Removing contradictory evidence from the summary

The conclusion appears cleaner but less reliable.

Extensive prose required for routine matters

Burden increases and generic language displaces material reasoning.

What good looks like

·    Another informed person can reconstruct the material reasoning.

·    The conclusion identifies decisive evidence and residual uncertainty.

·    Contradictions and alternatives have not disappeared without explanation.

·    The act of recording can trigger additional work when the chain is incomplete.

·    The amount of documentation is proportionate to significance and coordination need.

Common superficial imitations

·    A polished chronology of work with no evidence-to-conclusion link.

·    Generic phrases such as "appears reasonable" or "sufficient appropriate evidence obtained".

·    Long descriptions that conceal the decisive judgement.

·    Confidence asserted without explaining what it rests on.

Connections to other CLEAR moves

Record and Reflect tests whether the Clarify question was answered, whether the Look Beyond field was properly bounded, whether Evidence resolved the material distinction and whether Ask obtained the necessary knowledge. A gap should trigger recursion, not better prose.

Design contrasts

Weak design

Near miss

Stronger design

Why the difference matters

"No exceptions noted. Balance is reasonable."

"Procedures were completed and evidence supports the balance."

"The issue was whether year-end sales were recognised before control transferred. Dispatch and customer acceptance evidence resolved the timing for the selected high-risk items; two post-year-end deliveries led to an expanded population and adjustment. No material timing pattern remains."

The stronger version connects question, decisive evidence, updating and closure.

"Management's forecast is reasonable."

"Assumptions are within an acceptable range."

"Headroom depends primarily on renewal volume and delivery capacity. Signed renewals and current recruitment support the first-year forecast; the planned margin step-up remains the weakest link but a 3-point reduction does not remove headroom. Closure is therefore supported, subject to the disclosed sensitivity."

It makes dependencies, evidence, weakness and threshold visible.

Appendix

Activating thinking through language – design risks 

Designing for CLEAR-Thinking is not principally about telling auditors about CLEAR. It is about designing the language and conditions that make the required thinking more likely to occur at the point where the work and judgement are being formed.

This is what we mean by activating thinking: intentionally designing for the type of mental activity we want the auditor to engage in.

Language is one of the designer’s most powerful materials because words do more than transfer information. A question, heading, instruction or label directs attention. It suggests what matters, what is already assumed, what kind of answer is expected and what the person should do mentally next.

For example:

Wording

What it is likely to activate

Apply professional scepticism to management’s explanation.

A general intention to be sceptical, but little direction about what to think about or do.

What materially different business events could explain this pattern, and what evidence would distinguish between them?

Generation of alternatives, mental simulation and a search for discriminating evidence.

Obtain support for management’s explanation.

Confirmation of an account that has already been privileged.

What evidence would distinguish management’s explanation from the strongest credible alternative?

Comparison between live explanations and a more diagnostic evidence search.

The difference is not simply that one version is more detailed or better written. The wording sets different thinking in motion. A designer should therefore evaluate language by asking not only:

Is this grammatical and technically correct?

but also:

What is this wording likely to make the auditor notice, assume and do mentally?

This is especially important because language carries two design risks that are easy to overlook: ambiguity and filling in, and priming or bias through words and labels.

Ambiguity and filling in

Words do not carry a perfectly fixed meaning from one person to another. People make sense of language by connecting it to their existing knowledge and experience. This means a familiar word such as challenge, reasonable, understand, investigate, reflect or professional scepticism may activate a rich and specific set of experiences for an expert auditor, but something much thinner or materially different for a less experienced auditor.

That creates a particular design risk for experts. Because the designer already knows what good thinking looks like, they can unconsciously fill in what the wording leaves unsaid. For example, an experienced designer reading…

Assess whether the forecast is reasonable.

…may automatically think about customer demand, capacity, recruitment, margins, historical forecasting accuracy, dependencies between assumptions and evidence that would contradict the forecast. None of that is actually contained in the instruction. A less experienced auditor may instead interpret it as:

Check the model, compare assumptions with last year, confirm that the budget was approved and conclude whether it looks sensible.

The designer experiences the instruction as clear because their own expertise silently supplies the missing mental operations. This is why metacognition is an important design skill. The designer has to notice the difference between what the words actually provide and what their own expertise is adding to them.

The design response is not necessarily to add more words. Additional explanation can simply create more ambiguity and cognitive load. Instead, aim to make the critical cognitive operation explicit. Use concrete referents, diagnostic questions, contrasts and examples where they help align understanding. For example:

Assess the growth assumption

Becomes

What specific changes in customers, volume, price or capacity have to occur for the forecast growth to be achieved, and where should we already see evidence of those changes?

The second version reduces the amount the auditor has to invent about what “assess” means, while leaving the judgement itself with the auditor. A useful test for designers is:

How much am I filling in from my own experience that the user has not actually been given?

Where the answer is “a lot”, the scaffold may be relying on expertise that the intended user does not yet possess. Remember, scaffolding can’t replace knowledge; it just nudges the thinking that draws on prior knowledge. This is why connecting junior colleagues to the real business activities that numbers are proposed to represent is a critical part of the development.

Priming and bias: words change what becomes salient

The second risk is that language does not enter a neutral mind. Words and labels activate associations, expectations and prior experiences. Those associations influence what becomes psychologically foregrounded and therefore what people notice, search for and evaluate.

The underlying information can remain the same while a different label changes which characteristics become salient in people’s assessment. The design lesson is that apparently minor wording choices can influence judgement before the person has consciously decided how to approach the task. This matters directly in audit design.

·       Asking the auditor to “corroborate” primes a search for confirming evidence.

·       Asking “why no further work is required” makes closure the implied normal answer.

·       Pre-populating a conclusion with “no issues noted” establishes an expectation about what should be found.

By contrast, describing the observation first and leaving its explanation open makes a different form of reasoning easier:

·       Revenue increased by 18%, concentrated in the final six weeks, while receivables increased more quickly. What materially different business events could produce this combined pattern?

The objective is not to eliminate every influence from language; that is impossible. The objective is to choose language whose influence is congruent with the thinking we want to activate.

This means designers should not dismiss wording differences as “just semantics”. Defining a word elsewhere in the guidance does not necessarily remove the associations the word already activates when encountered in the flow of work. Use this effect deliberately. Choose labels, verbs and questions that make the appropriate object of attention salient: the audit question rather than the procedure, the observation rather than the preferred explanation, the distinction between explanations rather than the amount of evidence collected, and the reasoning behind the conclusion rather than the completion of the file.

A practical language-design test

For any important prompt, instruction, heading or field, mentally place yourself in the position of the intended auditor and ask:

1.     What do I notice first?

2.     What does the wording appear to treat as already true?

3.     What answer does it make feel expected or easiest?

4.     What mental operation does the key verb actually ask me to perform?

5.     What am I supplying from my own experience that may not be supplied for another auditor?

6.     Could someone produce a professional-looking response without carrying out the CLEAR-Thinking we intended?

If the answer to the last question is yes, the design is vulnerable to pseudo CLEAR-Thinking.

Supporting communications
materials