Part I shared some of Grant Purdy’s background and experience from working more than 40 years on practical applications of risk management, including the improvement of decision making. Part II focused on Grant’s new book on decision making that he co-authored with Roger Estall. Part III continues our conversation with Grant about his new book and how best to help “Deciders” make even better decisions.
Let’s turn to the various components of decision-making, starting with communication. Roger and you use the term “conversing.” What makes for a quality conversation and how is this helpful to decision-making?
Deciders talk with others because of their innate awareness of and ‘discomfort’ with uncertainty. This is part of the human condition. Whether intuitively or deliberately, Deciders seek to replace uncertainty with more certainty. Hence, when they turn to others, they are invariably either seeking confirmation, trying to fill gaps in their own knowledge and understanding, obtaining further perspectives, or searching for other options.
So yes, these interactions include ‘communication’ in the sense of the structured transfer of information, but much of the conversation that Deciders engage in, is more explorational - testing ideas, testing assumptions, testing motives - as well as acquiring information. The extent to which these conversations contribute to the Decider gaining sufficient certainty as to the actual outcomes that the decision will deliver (and therefore, the actual contribution to the organisation’s Purpose) has much to do with the quality of those conversations.
In our book ‘Deciding’ we suggest that there are three aspects of any conversation that will determine its quality and effect: its tone, how it is structured and, also, how various conversational techniques are deployed. The tone is what empowers and encourages the other participants in the conversation to engage with the question and contribute their thoughts as best they can. Tone influences perceptions as to the purpose of the conversation, and so serves to regulate the level of enthusiasm and engagement of participants. In our book, we contrast the likely effects of a ‘best idea wins’, and ‘there are no bad ideas’ - tone, with that of ‘my idea wins’ or a ‘who told you to speak?’ approaches.
Setting the tone isn’t a process. Tone is nuanced: it enables (or detracts from) a free flow of ideas and information directed to a common goal. When Roger Estall (my co-author) and I served on ISO standards-writing committees, we found that too many committee chairs would impose strict ‘speak in turn’ or ‘only one comment each’ rules which although, perhaps, encouraging participation, didn’t enable or encourage intellectually rigorous debates and engagement with and testing of each other’s thoughts and arguments.
Structuring the conversation –particularly when there are several participants - determines whether discussion proceeds logically, that no one’s time is wasted and that everyone is clear on why the conversation is being held and what the Decider needs from it. This also applies to one-on-one conversations. Furthermore, by consciously structuring the conversation, Deciders often bring more clarity to their own understanding of the question and what they are seeking to achieve from the conversation. Roger told me once of his first boss, who he described as someone who had a practical approach to practising religion. He told Roger of what he saw as the power of prayer, saying that in order to pray, one had first to get one’s thoughts in order so as not to seem stupid in conversation with God. And so it is with conversations in the context of decision-making.
Structure also requires consideration of participants – not only with respect to the span of their knowledge and expertise, but also to their likely contribution to tone. For example, it can be helpful to include ‘nay sayers’ (devil’s advocates) who have an instinct for challenging ‘group think, ‘futurists’ who are able to look ahead to envisage the consequence and possibilities of future change over the life of a decision, and people with ‘skin in the game’ – i.e. those who will be involved in implementing the decision.
Underpinning tone and structure are the techniques used by Deciders. In Deciding we mention several of the more influential techniques including those that:
Purpose and context are also important to effective decision-making. What do you mean by these terms and how do they impact decision-making?
As I hope is the case with all words throughout our book, these two words are used with their ordinary meanings. ‘Purpose’ is that for which the organisation exists and the reason, therefore, for making each decision; ‘context’ is the background which gives rise to opportunities and against which decisions to exploit those opportunities are made and implemented. Clarity and understanding of both are therefore vital to successful decision-making.
We preferred the word Purpose to ‘objective’ as the latter is used at different levels to label all manner of aspirations. By contrast if you ask people about the Purpose of their organisation, while they may not always express it the same way, they will all understand what the question means. From Purpose flow decisions that set objectives, strategies, and plans and hence it is the highest expression of the reason an organisation exists. Whether articulated or not, the Purpose reflects the values to which the organisation aspires and is the signpost, pointing the way to what it seeks to achieve.
This is why clarity of Purpose is a necessary starting point for Deciders and their related conversations. As I noted earlier, making sure that all involved in these conversations understand the Purpose – i.e. making sure they are all on the same page, as it were – is one of the important techniques for successful conversations. There is a side-benefit in always starting conversations this way: it soon builds an intuitive awareness of Purpose and reduces the potential for Deciders to overlook it or deviate.
Achieving a shared understanding of Purpose is a highly beneficial activity as, without it, Deciders and those involved in making and implementing decisions will inevitably substitute their own understanding. This in turn invariably leads to unintended outcomes that detract from rather than help realise Purpose.
In the book, we illustrate just how dangerous it can be for there not to be a shared understanding of Purpose within an organisation by reference to a recent major inquiry in Australia (via a ‘Royal Commission’) regarding the finance sector. This revealed that even the directors of some major financial institutions had deviated from their institution’s Purpose when making important decisions, for the simple reason that there was no common understanding and reinforcement of that Purpose. The resulting impacts on careers, reputations and financial performance proved costly.
Our many references to the significance of the ‘context’ in which decisions are made and then implemented, reflects the fact that decisions don’t exist as islands. Instead, decisions interact with their surroundings in terms of what is possible, what is valid, and what will be the actual outcomes.
In the chapter dedicated to context, we explain that context is specific to each decision and invariably has many aspects, albeit that many aspects may be relevant to other decisions. To provide a structure for thinking about context in an orderly way, we have talked about the internal, external and wider context with the latter band referring to those things over which the Decider or their organisation has little direct influence or control (e.g. the weather, social media and politics).
Apart from its various manifestations, the other notable aspect of context is that many of its elements are dynamic, and even more so, over time. Furthermore, there are often interactions between these elements which themselves can change independently of the elements. We illustrate the profound importance of context by reference to the phenomena of the sudden emergence of ride-sharing services such as Uber, which cleverly leveraged apparently unrelated changes in the context in which the regular cab industry operated (e.g., advances in GPS mapping, on-line payment systems, proliferation of smart phones) to build an entirely different and disruptive business model.
Deciders must thus not only consider the present, but for those decisions that will be implemented on an ongoing basis, the potential for future changes in context. That is because decisions invariably are founded on the validity of assumptions, many of which will relate to context. Deciders who appreciate the potential for changes in context to influence actual outcomes, will thus be likely to consider how such changes can be detected, understood, and responded to. Moreover, they will do this as a part of making the decision and as necessary, include secondary elements in the decision that monitor and respond to context change as required.
Many organisational failures and disruptions can be linked to either or both failure to detect or understand the significance of change in context, and being unprepared to respond to change. Nonetheless responding to these challenges is not always easy. While major change is usually quickly identified, the cumulative effects of minor or subtle change can be harder to recognise yet be equally important. As an analogy, we point out that death by the proverbial thousand cuts is just as final as death by execution and so deserves equal attention!
These days, macro data, generated and processed by artificial intelligence systems, can assist in the detection of gradual change. However, we also make the case for large organisations to employ people with the instincts to search for and recognise the significance of changes – especially those occurring in the wider context. Such people are sometimes described as futurists.
Assumptions - I feel like we could devote an entire book to this concept because so little in life is certain. How should we go about identifying and understanding assumptions?
I mentioned earlier that over our careers Roger and I have investigated many decisions that had gone spectacularly wrong and resulted in unintended outcomes. We almost always found (and I think it is generally accepted) that the most common root cause of failure, disaster and general ‘stuff-ups’ is a lack of awareness by the Decider of having made an assumption and/or misunderstanding its significance.
So the short answer to your question is for Deciders to focus on first enhancing their awareness of each assumption they make, and secondly exploring the twin issues of the importance of the assumption to achieving the desired outcome, and the level of certainty about the assumption. Together, importance and certainty define the significance of the assumption.
Let’s start with awareness.
Being aware of assumptions becomes much easier when Deciders develop an intuitive acceptance that the validity of each decision will, as a matter of inevitability, be based on one or more assumptions – assumptions that they have made. Even if the assumption has not been made consciously, if one knows it is there, it becomes more natural to actively search for what has been assumed. This is often best achieved in the conversations that take place around each decision.
Hence, in our book, we advocate those participating in decision-making conversations routinely ask ‘out-loud’ questions such as “okay…what are we assuming here?”, or “what have you assumed in reaching that conclusion?” or “why do we think this (or that) is so?” or “to what extent could this change over time?”. In other words, address the issue (or, as we call it, “the elephant in the room”) explicitly. The more experienced the Decider, the more they do this as a routine and conscious aspect of the iterative process through which they develop and evaluate decision options.
Understanding assumptions can be a bit more challenging but in any event, has two dimensions – understanding the source of the related uncertainty (i.e. “why aren’t we certain?”) and the significance of the assumption in terms of achieving the desired outcome and contributing to purpose (i.e. “why does it matter?”). The source of uncertainty is relevant because this provides a pointer to how, if beneficial to do so, the significance of an assumption can be reduced by reducing uncertainty.
Bias is a case in point. All people have biases. Deciders need to accept that this also applies to themselves and hence be curious as to how, and the extent to which, bias is feeding into assumptions and decisions by either disregarding or underestimating uncertainty. As we explain, biases have many dimensions with ‘confirmation bias’ (unjustified confidence), ‘conformity bias’ (group think) and ‘short term-ism’ (disregard of the future) being some of the more common.
If uncertainty is a consequence of a shortage of specific information then, clearly, further research or investigation can either reduce or eliminate this and replace it with greater certainty.
Uncertainty about the future (over the life of a decision) is trickier to address although can sometimes be at least reduced by understanding the past and attempting statistical projections of the future. Here, as already discussed, the most difficult changes to consider are those that relate to context – which is in itself a reason (especially for major decisions) to record the context as it was at the time the decision is made.
The significance of an assumption relating to uncertainty about the future, might also be able to be reduced by ongoing monitoring arrangements to detect relevant change with, perhaps, trigger points or ‘tripwire’ alerts, combined with contingency arrangements that will respond or can be activated if such change is detected. Stock market players use these defences all the time.
For a decision based on an assumption about the future and to provide sufficient certainty about the (future) outcome, these monitoring and contingency arrangements will need to be built into the decision itself as ‘secondary elements.’ I explained earlier that the primary elements of a decision are those that exploit the opportunity and the secondary elements are those that make it more certain that the desired outcomes will result.
In our book, we included a simple graphic to help Deciders think about the combined effects of the importance of an assumption (i.e., the extent to which the desired outcomes of the decision are dependent on the assumption, and the importance of those outcomes to the organisation’s Purpose) and confidence in the assumption. This graphic does not provide a formulaic approach (although some aspects of uncertainty may be quantifiable) but, rather, a consistent framework to use during conversations for considering assumptions that incorporates a simple, deliberately blurry-edged hierarchy to better inform where and whether to put the effort to reduce significance.
Our book also includes a template to help Deciders think about assumptions that relate to changes in context. It is a reminder that it is often not just the possibility of change that must be appreciated, but also the probability, velocity, timing and detectability of change, as these all influence the level of certainty, and, therefore, the suitability of monitoring and contingency arrangements. This template is intended to be conceptually informative, although it can also be populated for major decisions, where it is either desirable or necessary to record how assumptions were understood and addressed. For those decisions that take place each day as part of normal work, we have condensed these concepts into a pocket/wallet sized prompt card which can be used anywhere. This is illustrated in one of the appendices to our book and is suitable for replication.
You make the point about using monitoring to detect any change from what is assumed to what is occurring. How do we go about that, and are there general rules for when we should revisit a previously made decision?
As I’ve noted earlier, to achieve sufficient certainty about the outcomes of a decision that are dependent on assumptions about the future, the decision will probably need to incorporate a monitoring regime as one of its secondary elements. This is because:
I’ve already referred to the potential for there to be changes in the context over the life of a decision (and, therefore, why monitoring relevant aspects of context is valuable) but there is also a need to monitor assumptions about implementation – particularly if it is not the Decider who will be doing the implementing.
Monitoring should, therefore, seek to detect whether:
There are many monitoring techniques and their selection relates to the parameter that is being monitored (for example, monitoring of the behaviour of people requires different approaches to monitoring of devices) and to the nature of the changes that could occur. For example, if the anticipated frequency of change is low, it may be sufficient to monitor infrequently, or on a sampling basis. If the consequences of change are high, then a more continuous method of monitoring might be more appropriate.
Whatever the monitoring method selected it will likely have its own set of assumptions. For example, that a monitoring device such as a thermometer in a frozen food storage room will function faultlessly and, therefore, dual thermometers need not be fitted. It’s also necessary to consider any assumptions that relate to the intended responses to monitoring – for example, will the significance of what is detected be correctly understood and acted upon within the assumed timeframes?
And there are also the questions of efficiency and trade-offs mentioned above. For example, is it better to incur the cost of higher reliability equipment or use more highly trained staff (with high levels of certainty as to performance in each case) in exchange for a lower-cost regime of monitoring and supervision? As I have said above, this is a further reason to design monitoring arrangements as the decision is being made, so that the best balance can be struck. We discuss these essential conceptual issues in one of the main chapters of the book and provide considerable supporting detail (together with examples and anecdotes) in an appendix about designing monitoring.
The approaches I have outlined so far work well for freshly made decisions, but all organisations also function, in part at least, on the basis of on-going implementation of previous decisions which did not incorporate monitoring arrangements. Indeed, in some cases, the mere fact of such decisions having been earlier made, can only be inferred by observing what is currently happening.
As with contemporary decisions, historical decisions were also made in the setting of the context that existed at the time and so are just as vulnerable to change over time. The problem occurs when, either through loss of institutional memory through staff changes or the absence of records, no one knows the prevailing context when a historical decision was made or, therefore, the assumptions then. Often, when organisations are challenged why they do a certain thing in a certain way, they can only respond in terms of “it’s always be done like that!”
Without an appreciation of the original context and assumptions, it is difficult to later design a monitoring regime. Yet blissfully continuing to operate, unaware of the assumptions underpinning existing practice, is unlikely to provide sufficient certainty about outcomes.
It’s clearly impractical to try to revisit all historical decisions – even if they could be identified. Therefore, for at least the most important decisions and for any significant assumptions, we describe a couple of pragmatic and practical approaches to establishing monitoring. These involve:
It seems that our world is becoming increasingly intertwined and interdependent, which, in turn increases the level of uncertainty. How important are core values to good decision-making especially when we may not have a lot of information upon which to form assumptions. Two examples of situations where we have a paucity of information and experience upon which to form fact-based assumptions are climate change and the coronavirus pandemic.
An interesting question! Let me take these points in turn.
As I mentioned in Part 1 in my reference to the complexity of modern aircraft and nuclear submarines, as examples, there is no implicit relationship between complexity and uncertainty. It is fair to say, however that the greater the complexity (whether as a result of interdependencies etc. or technical complexity) the greater is the challenge of achieving sufficient certainty about the related decisions. The main point about this, is that it becomes even more important for decisions about complex issues to apply the universal method diligently (for example, around the conversations and transparency and consideration of assumptions and biases) but also to ensure the availability of expertise in relation to all elements of the decision.
The relationship between availability of relevant information and assumptions works in this way:
It is yet another reason why not only assumptions, but assessments of the level of certainty should be highly transparent. Too often – particularly on controversial issues – opinions or, even, findings (which are, after all just the product of decisions) are presented as if they were fundamental truths rating along with wind, fire, taxes, and gravity! This is even more so when modelling is used to bridge gaps in empirical facts - as has been the case with both the climate and Covid19 examples that you cite. Suitable models are often the only way we can inform decisions about situations we have either neither or rarely encountered before. However, all are based on assumptions and can contain biases and inherent uncertainties that must be made clear to Deciders.
Similar considerations apply to ‘values’ (core, or otherwise). I mentioned earlier in this interview in answer to your question about an organisation’s ‘Purpose’ that it referred to the highest expression of the reason for its existence and thus was likely to reflect the organisation’s values. But as with reasons for existence, values also vary both between organisations and over time. They are not like the relatively constant force of gravity with which they are sometimes likened.
So even though America’s Declaration of Independence from Britain asserts that “We hold these truths to be self-evident. . . ,” those truths (which were values), sadly, are not universally evident. While they might be substantially evident in the United States, they are not necessarily evident (or even aspired to) in other countries. So, while we might be tempted to believe that there are universal, or ‘core’ values, that is neither demonstrably so, nor can it be imagined - principally because there is no universal decision-making entity for establishing or selecting such values.
Which brings me to an important point about Deciders. For someone or something to be a Decider, they must possess the power to decide and this is a real challenge for global issues such as addressing changes in climate and, since the beginning of this year, Covid19. It is not just the adequacy of (and confidence in) available information that is problematic but the relatively weak institutions (without universal trust or power) that are available for making decisions in relation to these issues.
Both the climate and pandemic issues are good examples of the importance of being clear about Purpose. In our book we caution about starting with an answer and trying to fit it to a hitherto undefined problem – the common failing, sadly, of so many advocates of ‘risk management.’ While there is obviously room for debate about the technical details of both issues, climate change and Covid19 are particularly contentious because of the absence of any universal agreement on Purpose.
In the case of climate, even where there is a large degree of agreement on the cause and extent of change, there is no universal agreement about whether to simply adapt to the change or attempt to reverse it or to adopt some kind of combined approach. And with the pandemic, there is no general agreement on the point of balance between health considerations and those relating to economic and general well-being. Instead, decisions are being made without agreement on Purpose and thus there is no basis on which to determine their adequacy.
While gaining such clarity is obviously easier said than done (particularly in the context of the unpredictable and distorting influences of mainstream and social media, competing interests and the widespread politicization of these issues) it is probably this lack of shared Purpose, rather than available information, that is the greater challenge for both the Deciders who are having to address these issues and their constituents who have to understand and determine the adequacy of these decisions.
I’ve really enjoyed the challenge of responding to your excellent questions and I hope my responses cause your readers to become more aware of their own decisions and, maybe, seek to make even better ones in future.
Deciders talk with others because of their innate awareness of and ‘discomfort’ with uncertainty. This is part of the human condition. Whether intuitively or deliberately, Deciders seek to replace uncertainty with more certainty. Hence, when they turn to others, they are invariably either seeking confirmation, trying to fill gaps in their own knowledge and understanding, obtaining further perspectives, or searching for other options.
So yes, these interactions include ‘communication’ in the sense of the structured transfer of information, but much of the conversation that Deciders engage in, is more explorational - testing ideas, testing assumptions, testing motives - as well as acquiring information. The extent to which these conversations contribute to the Decider gaining sufficient certainty as to the actual outcomes that the decision will deliver (and therefore, the actual contribution to the organisation’s Purpose) has much to do with the quality of those conversations.
In our book ‘Deciding’ we suggest that there are three aspects of any conversation that will determine its quality and effect: its tone, how it is structured and, also, how various conversational techniques are deployed. The tone is what empowers and encourages the other participants in the conversation to engage with the question and contribute their thoughts as best they can. Tone influences perceptions as to the purpose of the conversation, and so serves to regulate the level of enthusiasm and engagement of participants. In our book, we contrast the likely effects of a ‘best idea wins’, and ‘there are no bad ideas’ - tone, with that of ‘my idea wins’ or a ‘who told you to speak?’ approaches.
Setting the tone isn’t a process. Tone is nuanced: it enables (or detracts from) a free flow of ideas and information directed to a common goal. When Roger Estall (my co-author) and I served on ISO standards-writing committees, we found that too many committee chairs would impose strict ‘speak in turn’ or ‘only one comment each’ rules which although, perhaps, encouraging participation, didn’t enable or encourage intellectually rigorous debates and engagement with and testing of each other’s thoughts and arguments.
Structuring the conversation –particularly when there are several participants - determines whether discussion proceeds logically, that no one’s time is wasted and that everyone is clear on why the conversation is being held and what the Decider needs from it. This also applies to one-on-one conversations. Furthermore, by consciously structuring the conversation, Deciders often bring more clarity to their own understanding of the question and what they are seeking to achieve from the conversation. Roger told me once of his first boss, who he described as someone who had a practical approach to practising religion. He told Roger of what he saw as the power of prayer, saying that in order to pray, one had first to get one’s thoughts in order so as not to seem stupid in conversation with God. And so it is with conversations in the context of decision-making.
Structure also requires consideration of participants – not only with respect to the span of their knowledge and expertise, but also to their likely contribution to tone. For example, it can be helpful to include ‘nay sayers’ (devil’s advocates) who have an instinct for challenging ‘group think, ‘futurists’ who are able to look ahead to envisage the consequence and possibilities of future change over the life of a decision, and people with ‘skin in the game’ – i.e. those who will be involved in implementing the decision.
Underpinning tone and structure are the techniques used by Deciders. In Deciding we mention several of the more influential techniques including those that:
- help ensure everyone is clear on the organisation’s Purpose (and on how the decision is to support that),
- create awareness of, and help counter various forms of bias,
- respond to differences in the confidence of participants to contribute their ideas without fear of recrimination, derision, or discomfort,
- support the ‘best idea wins’ tone, and
- encourage participants to test their own ideas.
Purpose and context are also important to effective decision-making. What do you mean by these terms and how do they impact decision-making?
As I hope is the case with all words throughout our book, these two words are used with their ordinary meanings. ‘Purpose’ is that for which the organisation exists and the reason, therefore, for making each decision; ‘context’ is the background which gives rise to opportunities and against which decisions to exploit those opportunities are made and implemented. Clarity and understanding of both are therefore vital to successful decision-making.
We preferred the word Purpose to ‘objective’ as the latter is used at different levels to label all manner of aspirations. By contrast if you ask people about the Purpose of their organisation, while they may not always express it the same way, they will all understand what the question means. From Purpose flow decisions that set objectives, strategies, and plans and hence it is the highest expression of the reason an organisation exists. Whether articulated or not, the Purpose reflects the values to which the organisation aspires and is the signpost, pointing the way to what it seeks to achieve.
This is why clarity of Purpose is a necessary starting point for Deciders and their related conversations. As I noted earlier, making sure that all involved in these conversations understand the Purpose – i.e. making sure they are all on the same page, as it were – is one of the important techniques for successful conversations. There is a side-benefit in always starting conversations this way: it soon builds an intuitive awareness of Purpose and reduces the potential for Deciders to overlook it or deviate.
Achieving a shared understanding of Purpose is a highly beneficial activity as, without it, Deciders and those involved in making and implementing decisions will inevitably substitute their own understanding. This in turn invariably leads to unintended outcomes that detract from rather than help realise Purpose.
In the book, we illustrate just how dangerous it can be for there not to be a shared understanding of Purpose within an organisation by reference to a recent major inquiry in Australia (via a ‘Royal Commission’) regarding the finance sector. This revealed that even the directors of some major financial institutions had deviated from their institution’s Purpose when making important decisions, for the simple reason that there was no common understanding and reinforcement of that Purpose. The resulting impacts on careers, reputations and financial performance proved costly.
Our many references to the significance of the ‘context’ in which decisions are made and then implemented, reflects the fact that decisions don’t exist as islands. Instead, decisions interact with their surroundings in terms of what is possible, what is valid, and what will be the actual outcomes.
In the chapter dedicated to context, we explain that context is specific to each decision and invariably has many aspects, albeit that many aspects may be relevant to other decisions. To provide a structure for thinking about context in an orderly way, we have talked about the internal, external and wider context with the latter band referring to those things over which the Decider or their organisation has little direct influence or control (e.g. the weather, social media and politics).
Apart from its various manifestations, the other notable aspect of context is that many of its elements are dynamic, and even more so, over time. Furthermore, there are often interactions between these elements which themselves can change independently of the elements. We illustrate the profound importance of context by reference to the phenomena of the sudden emergence of ride-sharing services such as Uber, which cleverly leveraged apparently unrelated changes in the context in which the regular cab industry operated (e.g., advances in GPS mapping, on-line payment systems, proliferation of smart phones) to build an entirely different and disruptive business model.
Deciders must thus not only consider the present, but for those decisions that will be implemented on an ongoing basis, the potential for future changes in context. That is because decisions invariably are founded on the validity of assumptions, many of which will relate to context. Deciders who appreciate the potential for changes in context to influence actual outcomes, will thus be likely to consider how such changes can be detected, understood, and responded to. Moreover, they will do this as a part of making the decision and as necessary, include secondary elements in the decision that monitor and respond to context change as required.
Many organisational failures and disruptions can be linked to either or both failure to detect or understand the significance of change in context, and being unprepared to respond to change. Nonetheless responding to these challenges is not always easy. While major change is usually quickly identified, the cumulative effects of minor or subtle change can be harder to recognise yet be equally important. As an analogy, we point out that death by the proverbial thousand cuts is just as final as death by execution and so deserves equal attention!
These days, macro data, generated and processed by artificial intelligence systems, can assist in the detection of gradual change. However, we also make the case for large organisations to employ people with the instincts to search for and recognise the significance of changes – especially those occurring in the wider context. Such people are sometimes described as futurists.
Assumptions - I feel like we could devote an entire book to this concept because so little in life is certain. How should we go about identifying and understanding assumptions?
I mentioned earlier that over our careers Roger and I have investigated many decisions that had gone spectacularly wrong and resulted in unintended outcomes. We almost always found (and I think it is generally accepted) that the most common root cause of failure, disaster and general ‘stuff-ups’ is a lack of awareness by the Decider of having made an assumption and/or misunderstanding its significance.
So the short answer to your question is for Deciders to focus on first enhancing their awareness of each assumption they make, and secondly exploring the twin issues of the importance of the assumption to achieving the desired outcome, and the level of certainty about the assumption. Together, importance and certainty define the significance of the assumption.
Let’s start with awareness.
Being aware of assumptions becomes much easier when Deciders develop an intuitive acceptance that the validity of each decision will, as a matter of inevitability, be based on one or more assumptions – assumptions that they have made. Even if the assumption has not been made consciously, if one knows it is there, it becomes more natural to actively search for what has been assumed. This is often best achieved in the conversations that take place around each decision.
Hence, in our book, we advocate those participating in decision-making conversations routinely ask ‘out-loud’ questions such as “okay…what are we assuming here?”, or “what have you assumed in reaching that conclusion?” or “why do we think this (or that) is so?” or “to what extent could this change over time?”. In other words, address the issue (or, as we call it, “the elephant in the room”) explicitly. The more experienced the Decider, the more they do this as a routine and conscious aspect of the iterative process through which they develop and evaluate decision options.
Understanding assumptions can be a bit more challenging but in any event, has two dimensions – understanding the source of the related uncertainty (i.e. “why aren’t we certain?”) and the significance of the assumption in terms of achieving the desired outcome and contributing to purpose (i.e. “why does it matter?”). The source of uncertainty is relevant because this provides a pointer to how, if beneficial to do so, the significance of an assumption can be reduced by reducing uncertainty.
Bias is a case in point. All people have biases. Deciders need to accept that this also applies to themselves and hence be curious as to how, and the extent to which, bias is feeding into assumptions and decisions by either disregarding or underestimating uncertainty. As we explain, biases have many dimensions with ‘confirmation bias’ (unjustified confidence), ‘conformity bias’ (group think) and ‘short term-ism’ (disregard of the future) being some of the more common.
If uncertainty is a consequence of a shortage of specific information then, clearly, further research or investigation can either reduce or eliminate this and replace it with greater certainty.
Uncertainty about the future (over the life of a decision) is trickier to address although can sometimes be at least reduced by understanding the past and attempting statistical projections of the future. Here, as already discussed, the most difficult changes to consider are those that relate to context – which is in itself a reason (especially for major decisions) to record the context as it was at the time the decision is made.
The significance of an assumption relating to uncertainty about the future, might also be able to be reduced by ongoing monitoring arrangements to detect relevant change with, perhaps, trigger points or ‘tripwire’ alerts, combined with contingency arrangements that will respond or can be activated if such change is detected. Stock market players use these defences all the time.
For a decision based on an assumption about the future and to provide sufficient certainty about the (future) outcome, these monitoring and contingency arrangements will need to be built into the decision itself as ‘secondary elements.’ I explained earlier that the primary elements of a decision are those that exploit the opportunity and the secondary elements are those that make it more certain that the desired outcomes will result.
In our book, we included a simple graphic to help Deciders think about the combined effects of the importance of an assumption (i.e., the extent to which the desired outcomes of the decision are dependent on the assumption, and the importance of those outcomes to the organisation’s Purpose) and confidence in the assumption. This graphic does not provide a formulaic approach (although some aspects of uncertainty may be quantifiable) but, rather, a consistent framework to use during conversations for considering assumptions that incorporates a simple, deliberately blurry-edged hierarchy to better inform where and whether to put the effort to reduce significance.
Our book also includes a template to help Deciders think about assumptions that relate to changes in context. It is a reminder that it is often not just the possibility of change that must be appreciated, but also the probability, velocity, timing and detectability of change, as these all influence the level of certainty, and, therefore, the suitability of monitoring and contingency arrangements. This template is intended to be conceptually informative, although it can also be populated for major decisions, where it is either desirable or necessary to record how assumptions were understood and addressed. For those decisions that take place each day as part of normal work, we have condensed these concepts into a pocket/wallet sized prompt card which can be used anywhere. This is illustrated in one of the appendices to our book and is suitable for replication.
You make the point about using monitoring to detect any change from what is assumed to what is occurring. How do we go about that, and are there general rules for when we should revisit a previously made decision?
As I’ve noted earlier, to achieve sufficient certainty about the outcomes of a decision that are dependent on assumptions about the future, the decision will probably need to incorporate a monitoring regime as one of its secondary elements. This is because:
- It’s when the decision is being made that the Decider has the greatest awareness and understanding of the assumptions and so is best placed to specify monitoring requirements.
- While contingency arrangements to respond to changes are also best made as part of the decision, to be of effect, the organisation needs to know whether and when to activate those arrangements if there is change from what has been assumed.
- There can be significant trade-offs between the costs of the primary and secondary elements of decisions - particularly because the costs of the latter include those associated with monitoring.
I’ve already referred to the potential for there to be changes in the context over the life of a decision (and, therefore, why monitoring relevant aspects of context is valuable) but there is also a need to monitor assumptions about implementation – particularly if it is not the Decider who will be doing the implementing.
Monitoring should, therefore, seek to detect whether:
- implementation of a decision occurs as assumed or intended;
- outcomes are those intended;
- over the life of a decision, changes in context have occurred that invalidate significant assumptions or mean the decision no longer provides the best response to the opportunity that it was intended to exploit.
There are many monitoring techniques and their selection relates to the parameter that is being monitored (for example, monitoring of the behaviour of people requires different approaches to monitoring of devices) and to the nature of the changes that could occur. For example, if the anticipated frequency of change is low, it may be sufficient to monitor infrequently, or on a sampling basis. If the consequences of change are high, then a more continuous method of monitoring might be more appropriate.
Whatever the monitoring method selected it will likely have its own set of assumptions. For example, that a monitoring device such as a thermometer in a frozen food storage room will function faultlessly and, therefore, dual thermometers need not be fitted. It’s also necessary to consider any assumptions that relate to the intended responses to monitoring – for example, will the significance of what is detected be correctly understood and acted upon within the assumed timeframes?
And there are also the questions of efficiency and trade-offs mentioned above. For example, is it better to incur the cost of higher reliability equipment or use more highly trained staff (with high levels of certainty as to performance in each case) in exchange for a lower-cost regime of monitoring and supervision? As I have said above, this is a further reason to design monitoring arrangements as the decision is being made, so that the best balance can be struck. We discuss these essential conceptual issues in one of the main chapters of the book and provide considerable supporting detail (together with examples and anecdotes) in an appendix about designing monitoring.
The approaches I have outlined so far work well for freshly made decisions, but all organisations also function, in part at least, on the basis of on-going implementation of previous decisions which did not incorporate monitoring arrangements. Indeed, in some cases, the mere fact of such decisions having been earlier made, can only be inferred by observing what is currently happening.
As with contemporary decisions, historical decisions were also made in the setting of the context that existed at the time and so are just as vulnerable to change over time. The problem occurs when, either through loss of institutional memory through staff changes or the absence of records, no one knows the prevailing context when a historical decision was made or, therefore, the assumptions then. Often, when organisations are challenged why they do a certain thing in a certain way, they can only respond in terms of “it’s always be done like that!”
Without an appreciation of the original context and assumptions, it is difficult to later design a monitoring regime. Yet blissfully continuing to operate, unaware of the assumptions underpinning existing practice, is unlikely to provide sufficient certainty about outcomes.
It’s clearly impractical to try to revisit all historical decisions – even if they could be identified. Therefore, for at least the most important decisions and for any significant assumptions, we describe a couple of pragmatic and practical approaches to establishing monitoring. These involve:
- attempting to recreate the historical decision by inference (like reverse engineering a piece of equipment for which the plans have been lost) and then use the universal method to determine whether it remains valid or requires revision or cancellation; or
- taking the view that continuing with present activities (and thus continuing to implement the related, albeit unknown, historical decisions) amounts to a new decision, and then making that decision in the usual way.
It seems that our world is becoming increasingly intertwined and interdependent, which, in turn increases the level of uncertainty. How important are core values to good decision-making especially when we may not have a lot of information upon which to form assumptions. Two examples of situations where we have a paucity of information and experience upon which to form fact-based assumptions are climate change and the coronavirus pandemic.
An interesting question! Let me take these points in turn.
As I mentioned in Part 1 in my reference to the complexity of modern aircraft and nuclear submarines, as examples, there is no implicit relationship between complexity and uncertainty. It is fair to say, however that the greater the complexity (whether as a result of interdependencies etc. or technical complexity) the greater is the challenge of achieving sufficient certainty about the related decisions. The main point about this, is that it becomes even more important for decisions about complex issues to apply the universal method diligently (for example, around the conversations and transparency and consideration of assumptions and biases) but also to ensure the availability of expertise in relation to all elements of the decision.
The relationship between availability of relevant information and assumptions works in this way:
- the less the information available the more that must be assumed and,
- in most cases, the more likely that the assumptions will have high significance because confidence in those assumptions will, invariably be low.
It is yet another reason why not only assumptions, but assessments of the level of certainty should be highly transparent. Too often – particularly on controversial issues – opinions or, even, findings (which are, after all just the product of decisions) are presented as if they were fundamental truths rating along with wind, fire, taxes, and gravity! This is even more so when modelling is used to bridge gaps in empirical facts - as has been the case with both the climate and Covid19 examples that you cite. Suitable models are often the only way we can inform decisions about situations we have either neither or rarely encountered before. However, all are based on assumptions and can contain biases and inherent uncertainties that must be made clear to Deciders.
Similar considerations apply to ‘values’ (core, or otherwise). I mentioned earlier in this interview in answer to your question about an organisation’s ‘Purpose’ that it referred to the highest expression of the reason for its existence and thus was likely to reflect the organisation’s values. But as with reasons for existence, values also vary both between organisations and over time. They are not like the relatively constant force of gravity with which they are sometimes likened.
So even though America’s Declaration of Independence from Britain asserts that “We hold these truths to be self-evident. . . ,” those truths (which were values), sadly, are not universally evident. While they might be substantially evident in the United States, they are not necessarily evident (or even aspired to) in other countries. So, while we might be tempted to believe that there are universal, or ‘core’ values, that is neither demonstrably so, nor can it be imagined - principally because there is no universal decision-making entity for establishing or selecting such values.
Which brings me to an important point about Deciders. For someone or something to be a Decider, they must possess the power to decide and this is a real challenge for global issues such as addressing changes in climate and, since the beginning of this year, Covid19. It is not just the adequacy of (and confidence in) available information that is problematic but the relatively weak institutions (without universal trust or power) that are available for making decisions in relation to these issues.
Both the climate and pandemic issues are good examples of the importance of being clear about Purpose. In our book we caution about starting with an answer and trying to fit it to a hitherto undefined problem – the common failing, sadly, of so many advocates of ‘risk management.’ While there is obviously room for debate about the technical details of both issues, climate change and Covid19 are particularly contentious because of the absence of any universal agreement on Purpose.
In the case of climate, even where there is a large degree of agreement on the cause and extent of change, there is no universal agreement about whether to simply adapt to the change or attempt to reverse it or to adopt some kind of combined approach. And with the pandemic, there is no general agreement on the point of balance between health considerations and those relating to economic and general well-being. Instead, decisions are being made without agreement on Purpose and thus there is no basis on which to determine their adequacy.
While gaining such clarity is obviously easier said than done (particularly in the context of the unpredictable and distorting influences of mainstream and social media, competing interests and the widespread politicization of these issues) it is probably this lack of shared Purpose, rather than available information, that is the greater challenge for both the Deciders who are having to address these issues and their constituents who have to understand and determine the adequacy of these decisions.
I’ve really enjoyed the challenge of responding to your excellent questions and I hope my responses cause your readers to become more aware of their own decisions and, maybe, seek to make even better ones in future.
RSS Feed