Ministers are not appointed because they are policy or portfolio experts. But to do their jobs well - ie. to make good decisions - they need to be able to interrogate, evaluate  and weigh the advice of experts.

The advice you receive is contestable and contested. It comes from different sources with different areas of expertise and different considerations (notably policy advice v political advice). 

The ability to ask better questions is therefore an essential skill. By the time you become a minister you’re probably pretty good at this already, having honed these skills as an MP, in committees and in your pre-politics career. But as a minister the level of skill required increases, because of the sheer volume of information you are required to make sense of, the fact that you are required to act on it by making decisions, and the fact that you are now accountable for outcomes and the use of public resources.

Amongst other things, asking better questions enables you to:

• Better understand and define problems

• See policy problems and opportunities  from multiple points of view, including competing perspectives and arguments.

• Identify and evaluate risks

• Monitor implementation

• Understand emerging issues

• Avoid group think

• Identify points of agreement and disagreement, so you can focus on opportunities to reconcile alternative viewpoints, broker compromise, and forge coalitions.

This page focuses on simple practical steps that you can take to ask questions more effectively. 

There are also several high-level frameworks that MKI teaches in Advanced Political Leadership and other courses which have a much broader range of applications, but can also be used to help you ask better questions.

Clear Leadership is highly relevant. It is an approach that provides a basis for productive personal interactions and .

It helps you engage with the ways people construct different meanings from the same situation. Amongst other things can help you understand why stakeholders reach different conclusions, uncover underlying assumptions and beliefs, and distinguish between interpretations and facts.

Adaptive Leadership promotes flexibility that enables your strategies and tactics to evolve as events and contexts change.

Adaptive Leadership foregrounds the role leaders play in shaping group dynamics, prompting questions around whose voice is missing and why, and questions that create enough discomfort to motivate change without overwhelming people.

Dragonfly Thinking enables rapid synthesis of various perspectives. Developed by ANU and MKI faculty member Anthea Roberts, island aided by AI it prompts you to view the problem from a series of different vantage points: different stakeholders, disciplines, timeframes, and assumptions.

A good question is one that elicits important information that would otherwise not have come to your attention.

As the Harvard Business Review puts it: The questions that get leaders and teams into trouble are often the ones they fail to ask.

The goal is to get yourself beyond your kneejerk reactions and initial preconceptions, your biases and blind spots, and to surface issues that might arise from people not in the room

COGNITIVE OVERLOAD

It’s easier to think of those questions when you slow down, identify and challenge your assumptions, and engage what Daniel Kahneman and Amos Tversky called “System Two” thinking: deliberative, effortful, analytical processes that enable you to evaluate and problem-solve rigorously, rather than rely on the gut feelings and snap judgements that govern most day-to-day decisions (“System One” thinking).

The volume of work, the speed at which responses are required, the media scrutiny and the stress that the high stakes of leadership creates render the time and headspace required for System Two thinking a rarity.

You can also think of this dynamic in terms of scarcity: 

…people experiencing any form of poverty (of  food, of love and company, of time or material resources) become intently  focused on their urgent needs…quite literally, scarcity  ‘captures the mind’. The mind’s focus on one thing steals cognitive and attentional ‘bandwidth’, impairing its capacity to perform other tasks, including  the ability to make well-reasoned judgments and to assess the long-term consequences of current decisions….chronically busy people, suffering from  a scarcity of time, demonstrate similarly impaired abilities and make self-defeating choices.

POWER DYNAMICS

The imbalance between ministers and those who advise them can also create barriers: ministers have power, but lack information; on the other side of the interaction, counsellors have information but lack power.

This can mean people come to the interaction with a set of unhelpful presumptions about you: what you consider important, what you’re really asking, how much bad news you’re willing to hear, what you already know or whether you’re willing to entertain countervailing views. 

For example, as one former deputy premier put it, “public servants often assume you only care about the politics”.

Public servants often don't know how much or how little to share without specific questions from the minister (and/or their office)- they gauge what they care about early from what initial questions are asked.

Presumptions aside, there is a natural incentive to tell you what you want to hear. There is some celebrity associated with being a minister, and people want to make you happy.

AGENDAS

Many of the people from whom you seek information also have a stake in the outcome of your decisions. This is unavoidable. It is in the nature of consultative decision making. But it does mean that people are incentivised to frame the data in different ways: to cherry pick, to push poll, to smuggle in assumptions, preferred outcomes and so on.

If you want to delve deeper into the art of navigating stakeholder engagement, the materials on Partnering Well are useful <insert link>.

SAVING FACE

This is an aspect of the power dynamic issue, but one that merits examination in its own right: sometimes ministers don't feel safe to admit they don’t know things. They worry it will make them lose face.

There is a deeply ingrained association in our culture between knowledge, intelligence and respect: if you don’t know something important, you can and do fall in some people’s estimation, undermining your  authority. This isn’t just about ego. The relentless media search for “gotcha” moments doesn’t help. 

Part of your task as a leader is  to push back against this narrative. By admitting when you don’t know something, or that you need further information, you create permission for others to do the same, which leads to clearer communication, surfaces issues and helps avoid groupthink.

In those situations where you genuinely fear the consequences of admitting to a lack of knowledge, consider having one-on-one or smaller groups with people you trust, especially before going into a larger discussion. Hypothetical style, open ended questions that take care to avoid assumptions can also be helpful.

This page treats questions primarily as a means of  addressing knowledge gaps. However questions are also a powerful tool for building trusted relationships and shaping culture. 

BUILDING RELATIONSHIPS

Research since the 1970s shows conversations serve two purposes: information exchange and impression management (Brooks & John 2018, p5) Follow-up questions possess particular power, signaling genuine interest and making conversation partners feel respected and heard. When ministers dig deeper into responses, they demonstrate engagement that strengthens working relationships.

ASSESSING YOUR INTERLOCUTORS

The way someone approaches a question can reveal their priorities, expertise, values, and assumptions about what you want to hear. This makes questioning a diagnostic tool for understanding team dynamics, departmental cultures, and individual motivations. 

CREATING PSYCHOLOGICAL SAFETY

Questions like "What went wrong and whose fault is it?" create blame-focused environments, while "What can we learn from this and how do we stop it happening again?" foster cultures of continuous improvement that encourages staff to surface problems early rather than hide them. 

There are strong connections here with the Learning Mindset that sits at the centre of the MKI Learning Model <cross link>: a culture where questions are asked to cast judgment, assign blame and as a means of exerting power is very different from a culture where questions are asked to learn.

SIGNALLING PRIORITIES AND CREATING ACCOUNTABILITY

When a minister consistently asks about climate targets or healthcare wait times, departments quickly recognise these areas demand attention and resources. Regular inquiries about specific metrics or outcomes signal that performance will be monitored, creating natural pressure for results without micromanagement.

Using questions this way requires care lest you create unintended consequences. It can, for example, drive behaviours that lead departments to focus on improving the metric rather than addressing the underlying issues. If all you care about is reducing the average hospital wait time, deprioritising serious operations that require long theatre and recovery times and prioritising short operations that people can recover from at home will drive that number down. Where people focus only on metrics it can create a 'perform at all cost' mentality where problems are hidden, and shortcuts taken (or numbers are just fudged).

The things that you signal are  important will have resources diverted to them. Be careful what you wish for.

In some ways the precise formulations of the questions you ask are less important than the environment in which you ask them. If you have created a culture of psychological safety, where people feel comfortable challenging established wisdom, sharing bad news, and admitting ignorance, your questions will elicit more useful responses than they will if you ask them in a  climate of fear or defensiveness. 

A powerful tool for improving the way you interact with others as a leader is the Clear Leadership framework <insert link>. Clear Leadership focuses on clearing out the “interpersonal mush” - the sensemaking, unspoken anxieties and unhelpful assumptions - that can inhibit clear communication. It encourages leaders to distinguish between what they actually know and the wants, thoughts and feelings that can colour that knowledge. By adopting Clear Leadership practices yourself, by modelling them for the people around you, and by encouraging others to adopt them, you can create a hospitable environment for collaboration.

CREATING A COLLABORATIVE ATMOSPHERE

In his 2022 review of Queensland’s integrity frameworks, Let the Sunshine in, Peter Coaldrake characterises the dynamic that can naturally arise between ministers and the public service as follows:

There is a view, repeatedly confirmed, that public service advice is too often shaped to suit what are assumed to be the preconceptions of the people receiving it, that the price for frank and fearless advice can be too high, sometimes devastatingly so, and the rewards too low. All this encourages a reluctance to depart from what is perceived to be the ‘official line’. (Coaldrake 2022, p9)

There are a number of forces that can drive this behaviour and make collaboration between you and the people around you challenging:

• The natural power dynamic. Government - both the public service and the political class - is hierarchical by design and hierarchy creates distance. People - particularly more junior staff - will be intimidated. 

• The (accurate) perception that your bandwidth is severely constrained. This drives people to be concise in their communication, which is generally a good thing, but can create unintended consequences: stage fright,  loss of nuance and detail, and narrow discussion framings that are intended to keep things moving along quickly rather than expansive frames that promote discussion.

• “Small p” politics as people seek status and power through their proximity to the minister, and shutting out or undermining rivals

• Chain of command can mean that people aren’t authorised to tell you certain things (or may feel that they aren’t)

As discussed elsewhere <link to relevant section of Getting Started with the Dept>, chain of command exists for good reasons and you should think carefully about risks and benefits before you disrupt it.  The other factors in the list are things you will  need to actively work against if you wish to create open dialogue.

Signal it explicitly: Tell your teams you want an open discussion, not an echo chamber. Give them permission to challenge. Position debate and discussion as part of What Good Looks Like - the vital opportunity to canvas the issues behind closed doors before they are subjected to the scrutiny of the wider world.

Set the terms of the discussion: Whilst you want participants to feel comfortable voicing a wide variety of viewpoints, you also don’t want to waste people’s time. Perhaps certain options are politically off the table, or a project is already some way down the road. In that case, there’s little point investing time in canvassing the impossible, and you should define the guardrails in setting up the discussion.

Be clear about the goals as well as the scope: what’s in; what’s out.  Do you want a recommendation or an open discussion? Is there a decision to be made or is the decision clear and you are working out how to progress it?

A former senior leader at the Department of Prime Minister and Cabinet offers the following example:

“The best I've seen this done is when a PM said he would be clear in all interactions on what he was seeking - advice, discussion or a clear decision has been taken and they wanted a way forward. This was then consistently backed up in all interactions to the point where he would say so at the start of a meeting "I'm seeking views and different ideas" vs "I'm clear on the outcome, I need a way to deliver it".

There is a fuller discussion on useful distinctions you can make to be sure you get what you want in the page on commissioning advice <insert link>.

Your manner matters: It’s difficult to be frank with someone who is unapproachable: “People are more forthcoming when you ask questions in a casual way, rather than in a buttoned-up, official tone.”

No matter what you say explicitly about the style of interaction you prefer, people will take their cues from the way you talk and act.  How you respond when given bad or unwanted news for the first time, or the first time you are directly challenged will shape what people do with future bad news - even if they weren’t present to witness it. The received wisdom about how best to work with you forms quickly. Once entrenched it is difficult to change.

Think about things like how formal your language is, where you sit in the room, your body language. There is a certain level of respect that attaches to your position, and you don’t have to try and be everyone’s best friend, but making the effort to demonstrate that you’re not trying to intimidate will be appreciated.

Show that you’re interested: This works on two levels: people will be more forthcoming when they understand that you are trying to satisfy your own understanding, rather than poke holes in what they’re saying or catch them out. And it helps build the relationship: most of us warm to people who are interested in what we do.

WHAT IS A GOOD ANSWER?

Be clear about what a good answer looks like. This goes back to making sure your teams know your preferences. Former Deputy Premier of Victoria John Thwaites suggests:

• Tell them you want to be fully briefed. 

• Tell them that you want to hear the bad news first - and that you don’t want to learn about it by reading it in the press

• Ask them to identify their assumptions

• Ask them to identify the sources of their evidence (“how do we know this?”)

After you have asked these questions once or twice, they will likely be addressed up front in any advice you receive.

GIVE NOTICE

In all but the most unusual situations, you’re not trying to catch people out. Ideally, you’re working with them cooperatively to try and solve problems together. 

There is little benefit, then, in surprising them with questions. Letting departments know what you plan to ask before you meet with them gives them time to ensure they have gathered the relevant  information.

It also builds trust, because they know you’re being transparent with them, and helps them anticipate your concerns in the future.

WHO ARE YOU ASKING?

Many a policy failure has its roots in a minister’s drift away from listening to a broad range of advice, preferring instead a few trusted advisors that very easily become an echo chamber.

In an examination of the ways that policy failures are studied, leading academics in the field of public administration Mark Bovens and Paul t’Hart note:

The biggest fiascos are not caused by division, ceaseless debate, all too powerful checks and balances and institutional paralysis, but by the closing-up of policy-making processes: concentrating authority in too few hands; constraining the scope and duration of deliberation; and shutting down diversity and dissent. (Bovens and t’Hart 2016, p663)

Just because people have different points of view doesn’t mean that anyone is trying to mislead you. People have different information available to them, different interests and different incentives, and this naturally leads them to have different perceptions of what is happening. Trying to understand these different accounts of reality - not just what they are, but why people hold them and why they differ from each other, can help you build a deeper, more comprehensive understanding  of what’s happening

As a non-expert, one of the most reliable ways for you to assess the quality of the information coming your way is triangulation: testing what you hear from your main advice streams (in most cases the department) against other sources. These can include:

• Agency heads in your portfolio: They view things from a different vantage point from the department. Group discussions that include department officials and agency heads can be fruitful

• Other stakeholders in your policy area: There will likely be a lot of these and you will need to use some judgement as to who to consult. One pragmatic rule of thumb is to prioritise those who will talk to the media, one way or the other. Your office should be gaining these viewpoints constantly to feed into any policy considerations throughout.

  This is another area where the Clear Leadership framework can be useful, because it provides ways of classifying stakeholders and thinking about the most effective approach to each group - the way you approach fence-sitters will differ from the way you approach allies, which will differ from the way you approach bad faith or enemy actors.

• Experts: Researchers, academics, people from think tanks

• The media: They don’t always get things right, but if there is a discrepancy between what you’re hearing from the department and what you’re reading in the paper, you’ll want to understand why.

• Ministerial colleagues: The other tremendously useful source of information is other ministers - particularly those that have held the portfolio previously (both still in the ministry and those that have left). 

  You don’t just want their views of what ought to happen; you also want their view of the facts, because the people they are talking to and the pressures bearing on them could lead them to a very different view of what is happening than the one you have.

  This is true even if you are a junior minister and the colleague whose advice you seek is relatively senior. As former Victorian Deputy Premier John Thwaites says:

I think junior ministers would be surprised. If they went to…maybe not the Premier, but certainly the Treasurer and other senior ministers, and said, “Look, I'd like to have a chat about this issue, which is causing me some concern, or whatever, they'll get the chat. A lot of times, people hang back more than they need to,  especially if they’re not just asking for something. Treasurers and Premiers are used to ministers coming to them with an ask. But if you're coming to them with a more of a strategic discussion, saying “Give me your advice”, I think you'll find that some senior ministers are prepared to do that.

This can be something your office takes carriage of through its relationships with other ministerial offices, because managing those relationships is part of its job. Or it can be something you do yourself, directly - just be clear about who is speaking to whom. 

CONSIDER THE GROUP DYNAMIC

Some discussions are more fruitful if you have them in a group rather  than one on one. For conversations about risk, problem definition and early stage project planning - anything where you want to turn the issues around from multiple points of view -  the whole is greater than the sum of the parts. People will bounce off each other and synthesise thinking in a way that they won’t if you have a series of parallel discussions.

Who should be in the room?

The value of group discussions comes from diverse perspectives. Eight policy people will not often add much to each other (especially if they all come from the same team). Including frontline staff or managers, regulators, people with a regional viewpoint etc will add much more. If you can’t get the full range of perspectives you would like in the room, you can still generate surprisingly useful insights by asking participants to put themselves in other stakeholders’ positions: How do you think a nurse would react to this? What would a farmer’s concerns be? How is this going to change the way people in the call centre do their jobs? 

John Thwaites’ experience suggests that the optimal number of people for a discussion of this type seems to be 6-8: enough to ensure there are no passengers, not so many that the conversation becomes unmanageable, or so crowded that people stop listening and are simply waiting for their turn to speak.

How should you contribute?

This goes back to the discussion of power dynamics above. Even if in your own mind you are simply another participant in the conversation, people are inevitably looking to you for signals.  

On one hand, people expect you to lead. On the other, any opinion you express has a gravitational force - people will be more inclined to agree with it, less inclined to voice dissent, and may attempt to moderate their views to avoid giving the impression of challenging you. 

It may be helpful to think of yourself more as a facilitator of the discussion, directing the flow of the conversation and drawing participants in at the right time, only voicing an opinion when you need to move the discussion along or something is headed off track.

One thing worth keeping an ear out for is what people aren’t saying. Often politeness or the need to maintain relationships will keep them from voicing risks or issues that challenge or cut across someone else's role, especially where there are power dynamics in the room.

HAVE A SYSTEM TO FLAG THINGS FOR FOLLOW UP

This doesn’t have to be elaborate. John Thwaites takes notes longhand, leaving a big margin at the side of the page where he notes anything he wants to chase up. Then whenever he returns to his notes to brief himself the next time he needs to engage with the matter, he can quickly get to grips with what he needs to move forward.

COUNTERACTING COGNITIVE BIASES

Cognitive biases are systematic errors in thinking that affect our decisions and judgments. These mental shortcuts, while often useful in daily life, can lead us astray when we need to make complex, high-stakes decisions. They can prevent you  from asking the right questions, considering alternative perspectives, or accurately weighing evidence.

The challenge of cognitive biases is particularly acute in political environments. The intense time pressure, enormous workloads, and constant public scrutiny make rational, deliberative thinking more difficult. The high-stress nature of political life, combined with the need to make rapid decisions on complex issues, creates fertile ground for biased thinking. 

Overcoming cognitive biases is inherently difficult because they operate largely below conscious awareness. They feel natural and correct, making it challenging to recognise when we're falling prey to them. However, with deliberate strategies and systematic approaches, politicians can develop tools to counteract these biases when evaluating policy advice.

Success requires recognising that cognitive biases are universal human tendencies, not personal failings. 

Optimism Bias

Optimism bias leads us to overestimate the likelihood of positive outcomes while underestimating potential risks and negative consequences. This bias can result in overly rosy projections about policy effectiveness or timeline feasibility. Politicians may also underestimate implementation challenges or public resistance. 

The Learning from Failure report, an independent review of Government processes for the development and implementation of large public programmes and projects, identifies over-optimism as a persistent factor in implementation failures (Shergold 2015, pp ii, 13, 26)

Countermeasures

• Institute "pre-mortem" analysis: Before making a decision, imagine it has failed and systematically work backward to identify what could go wrong

• Require advisers to present both best-case and worst-case scenarios for proposals

• Establish devil's advocate/ black hat roles in advisory teams to challenge optimistic assumptions

Loss Aversion

Loss aversion causes people to weigh potential losses more heavily than gains, leading to excessive risk avoidance or clinging to failing policies. Politicians may resist necessary policy changes because they focus disproportionately on what might be lost rather than what could be gained.

Countermeasures

• Reframe policy discussions in terms of opportunities and gains rather than just costs and risks

• Be deliberate in creating your frame of reference by comparing current situations to worse alternatives, not just better ones

• Implement regular policy reviews that treat continuation as an active choice requiring justification, not a default

Anchoring

Anchoring occurs when initial information disproportionately influences subsequent judgments. The first statistics presented, expert opinions heard, or policy options considered can unduly shape final decisions. This is particularly problematic when advisers present information in ways that prime certain conclusions.

You can of course use anchoring to your advantage. It is one of the most powerful ways of shaping perceptions, and one of the most difficult to counter - think about the importance media advisors place on “getting ahead of the story" to control the narrative. It’s an extremely potent negotiation tool, implicitly setting the terms of the discussion..

Debiasing requires the deliberate pursuit of credible alternative accounts, something that requires significant investment.

Countermeasures

• Resist forming an initial view. Suspend judgment and seek multiple independent sources of information before locking into a problem definition os a particular solution.

• Deliberately generate alternative policy options before evaluating any single proposal

• Use structured decision-making processes that consider multiple criteria systematically rather than relying on first impressions

Sunk Cost Fallacy

The sunk cost fallacy leads people to continue failed courses of action because of previous investments -  time, money or - in ministers’ cases - political capital. The more resources already committed, the harder it becomes to change course, even when evidence clearly indicates the policy isn't working.

Of all the cognitive biases, sunk cost fallacy is probably the most challenging for politicians to rise above, because acknowledging failure, correcting course or abandoning a project all tend to create bad optics, handing your opponents a victory.

The challenge then is twofold: recognising one has fallen prey to the bias, and minimising the political damage of walking it back.

Countermeasures

• Create regular review cycles that evaluate policies based solely on future costs and benefits, not past investments. Political considerations may prevent you from acting on these evaluations, but at least you will be able to make a clear-eyed assessment of the trade offs.

• If you can, present the policy up front in a way that allows for refinement: as testing-and-learning, rather than the perfect solution.

• Ensure that monitoring and evaluation is ongoing, to allow for continuous improvement/ Consider debiasing by separating implementation decisions from evaluation decisions by using different teams or advisers for each

The Availability Heuristic

The availability heuristic causes people to overweight easily recalled information, often leading to policy decisions based on recent, memorable, or emotionally vivid events rather than comprehensive evidence. Politicians may overreact to crisis situations or media coverage while neglecting less visible but more significant problems.

Countermeasures

• Require comprehensive data analysis that includes base rates and historical context, not just recent examples

• Use structured checklists that ensure consideration of different types of evidence and stakeholder perspectives

• Consider implementing waiting periods for major policy decisions following high-profile events to allow for more balanced analysis. Careful though - the policy here can be in tension with the politics, because high-profile events can generate a window of opportunity that may not last. Gun reform, for example, might never have gotten up if Howard waited. Some states just would not have agreed three weeks later. This got up because all of the work had been done and it was basically ready to go.

Confirmation Bias

Human beings naturally seek information that supports their existing beliefs while avoiding contradictory evidence. This bias can lead to echo chambers where advisers tell leaders what they want to hear rather than what they need to know.

Countermeasures

• Actively seek out credible opposing viewpoints and require advisers to present them fairly

• Create formal processes for red-team analysis where advisers argue against preferred policies

• Establish relationships with trusted advisers who are explicitly empowered to challenge assumptions and provide dissenting views

TACTICS, TIPS AND RULES OF THUMB

Prefer Open Questions

Open questions are those that cannot be answered with a simple "yes," "no," or single-word response. They invite elaboration and explanation, yielding richer, more comprehensive information than closed alternatives.

They are necessary, but not sufficient, for avoiding incorrect assumptions, encouraging people to raise issues they consider important rather than limiting them to your predetermined concerns. This makes them invaluable for ministers seeking to understand complex policy landscapes or organisational challenges.

Open questions are particularly useful when interrogating processes, which tends to attract closed questions (“Do we have a risk register?”, “Do we have an implementation plan?”).

You will likely elicit more useful information if you ask questions that go to the content of these processes: “If this fails, why will it fail?”, “What are the implementation milestones? Why those? What could stop us from reaching them?” 

Ask what has been learned from particular aspects of the planning. “What feedback have you heard from frontline staff/teachers/nurses? What are their biggest concerns?” That will tell you more about the effectiveness of engagement than just asking if it has been done will.

Use Closed Questions Selectively

There is a role for more tightly framed questions, for example when you want to clarify an ambiguity, “...if you are in a tense negotiation or are dealing with people who tend to keep their cards close to their chest…. In such situations “...open-ended questions can leave too much wiggle room, inviting [respondents]  to dodge or lie by omission”

Research indicates that respondents are likely to be more forthcoming, and less likely to evade or prevaricate, if you frame your question using pessimistic assumptions. So “We have concerns about the capacity of the service providers to meet the Service Level Agreement, yes?” rather than “..and the service providers can hit those targets?”

Think about Question Sequence

If you know that there are going to be difficult or awkward questions, research indicates that it may be more effective to get these out of the way first, rather than building up to them as you might naturally want to do:

During tense encounters, asking tough questions first, even if it feels socially awkward to do so, can make your conversational partner more willing to open up…people are more willing to reveal sensitive information when questions are asked in a decreasing order of intrusiveness. 

There is a caveat here: this approach works best in the context of an established relationship:

If the goal is to build relationships, the opposite approach—opening with less sensitive questions and escalating slowly—seems to be most effective. 

This is why journalists are taught to start interviews with easier, less threatening questions before moving to sensitive topics. Active listening - reflecting back what someone has said with phrases like "So what you're saying is..." - shows you're engaged and gives sources a chance to clarify or expand.

Methodically Identify Your Assumptions

Assumptions, famously, ‘make an ass of u and me’. 

History furnishes us with a rich list of examples where incorrect assumptions led to disaster: the unexpected effects of low temperatures on the O-rings on the space shuttle Challenger; the Maginot Line defences that France constructed along its borders in the lead up to the Second World War, and which the German forces simply went around; and the shortage of lifeboats on the “unsinkable” Titanic.

Hindsight is a wonderful thing. The challenge is that assumptions don’t always identify themselves as such. They often masquerade as facts, self-evident and nearly invisible to conscious examination. 

You can't interrogate every assumption - that leads to paralysis. But you can integrate some practices into your ways of working that minimise the risk of being caught out.

The management researcher Sydney Finkelstein advocates a simple habit: 

Methodically revisit your assumptions…You can avoid…errors by regularly surfacing and testing your ingrained ideas. At the start of a new project or assignment, jot down three or more “theories” that underpin it. Analyze these assumptions one by one; decide which are valid and which you should discard; and change your strategy or approach accordingly. (Finkelstein 2019, p 155)

The Five Whys

The "5 Whys" is a root cause analysis method developed by Toyota founder Sakichi Toyoda in the 1930s and later refined as part of the Toyota Production System. The technique involves asking "why" five times in succession to drill down from a surface problem to its underlying cause.

A classic Toyota example: A machine stops working. Why? A fuse blew. Why? The bearing was overloaded. Why? It wasn't lubricated properly. Why? The lubrication pump wasn't working. Why? The pump's intake was clogged with metal shavings. The real problem wasn't the machine failure but inadequate filtration systems.

The method works because it forces systematic thinking beyond immediate symptoms. Each "why" peels back another layer, preventing the quick fixes that often address effects rather than causes. It's simple enough for anyone to use without special training, making it democratically applicable across an organization.

The technique has been adapted widely beyond manufacturing - in healthcare for patient safety analysis, in software development for debugging, in journalism for investigative reporting, and in personal problem-solving. The core principle translates: keep asking why until you reach actionable root causes.

Useful General Questions

There are all sorts of reasons people might not be telling you what you need to know. Perhaps they lack the context to appreciate the magnitude of the issue. Maybe they’re hoping it will resolve itself before they have to escalate it. Or perhaps they are hoping they can fix it themselves.

In these cases you can encourage people to share their concerns by framing questions that indicate the bar for raising an issue is low, and that no blame attaches - you’re not implying that they are covering anything up.

Such questions include:

• What keeps you awake at night?

• Is there anything else I should know?

• Is there anything that might become an issue?

• What am I not asking you that I should?

• If in two years’ time, this has failed, why?

• In 12 months, what is the negative story in the media? Or what is the positive story in the media?

• What do you need to believe to be true for this to succeed?

• If <x assumption/belief> was different would you make a different decision?

• Why has this not been done before? Or why did it not succeed in the past?

• If we couldn't do this, what would be the next best thing?

Develop a List of Red Flags

Former Prime Ministerial Chief of Staff Raj Nahna shares that when an agency or a department described something as “a technical solution”, alarm bells would start ringing in his head. His experience suggested that the phrase indicated not only that the speaker was trying to escape scrutiny by positioning their proposal as detailed and requiring specialist knowledge to understand, but that they were thinking in siloes, and not considering the wider implications.

For John Thwaites, it wasn’t so much a particular form of words that put him on high alert, it was what previous interactions about the character of the speaker. Thwaites is an advocate of triangulation - consulting multiple sources of information to ensure you have examined an issue from all angles. This process can sometimes reveal that a particular source has predictable tendencies - perhaps a willingness to gild the lily, a blindspot, or a habit of omitting information that does not reflect well on them.