top of page

The Wisdom Gap: Better Project Data Does Not Guarantee Better Decisions

Nov 9
7 min read

By James Bawtree




Projects have never produced more data. Executives can access dashboards showing schedules, expenditure, risks, benefits, resources and dependencies. Digital platforms can combine information across entire portfolios. AI can review thousands of documents, identify patterns, summarise competing evidence and forecast what may happen next. Yet better information does not guarantee better decisions.  


A project can have accurate data and still pursue the wrong outcome. A portfolio can be highly optimised while directing resources away from those who need them most. A decision can be technically defensible while creating consequences that leaders have failed to consider. This is the wisdom gap: the distance between knowing more and deciding well.  


As AI becomes more capable, the gap may become more significant. AI can help leaders understand what is happening, what is likely to happen and which options appear most efficient. It cannot independently determine what ought to matter. That remains a human responsibility. 


The Wisdom Gap: Better Project Data Does Not Guarantee Better Decisions

Data can inform a decision without resolving it 


Many project decisions appear technical but contain underlying value judgments. A schedule acceleration may reduce time to market while increasing pressure on employees and suppliers. A lower-cost design may meet minimum requirements while reducing resilience or accessibility. A digital service may improve efficiency for most customers while excluding people who cannot easily use it. 


Data can help leaders understand these consequences. It cannot decide which consequences are acceptable. Consider a major infrastructure project facing a choice between three delivery options. One is cheaper, another can be completed sooner and the third offers greater long-term environmental and community value. There is no neutral formula that can determine the correct choice. The outcome depends on how leaders value affordability, urgency, resilience, sustainability and fairness. It also depends on whose interests are represented and which impacts are included within the analysis. Every decision model gives weight to some things and less weight to others. The weights may be explicit, as they are in structured multi-criteria analysis, or they may be hidden within assumptions, business-case templates and historical data. Either way, the decision reflects values. 


project data analysis
project data analysis

Optimisation can conceal the real choice 


AI is particularly effective at optimisation. Given an objective, constraints and sufficient data, it can compare options more quickly than a human team. The danger lies in assuming that the objective itself is neutral. If AI is asked to identify the lowest-cost delivery model, it can optimise for cost. If it is asked to reduce schedule duration, it can recommend ways to accelerate delivery. If it is asked to maximise financial return, it can compare investments accordingly. But who decided that cost, speed or financial return should dominate? 


The most important decision may have occurred before the analysis began, when someone selected the objective and determined which impacts would be measured. This is why apparently technical decisions must still be examined as governance decisions. An efficient recommendation may be efficient only because some costs have been transferred to employees, communities, future asset owners or the environment. AI may not conceal these consequences deliberately. They may simply be absent from the data or the question it has been asked. The quality of an AI-supported decision therefore depends on more than the quality of the model. It depends on the wisdom used to frame the problem. 


team collaboration session
team collaboration session

Intelligence and wisdom are not the same

 

Intelligence helps us process information, recognise patterns and solve problems. Wisdom helps us determine which problems are worth solving, which outcomes are desirable and what should be done when legitimate values conflict. In project governance, wisdom involves understanding the system around the project. It requires leaders to consider immediate outputs alongside long-term consequences, intended benefits alongside potential harm, and organisational objectives alongside wider public value. This does not mean relying on intuition instead of evidence. Wisdom is not the rejection of analysis. It is the responsible interpretation of evidence within a broader understanding of purpose, values and consequences. A wise decision may still involve compromise. It may impose costs on some stakeholders or accept a level of residual risk. The difference is that these trade-offs are recognised, examined and owned rather than hidden behind a model or dashboard. 


team discussion meeting
team discussion meeting

Values-thinking as a project leadership competency 


Sustainability research identifies values-thinking, sometimes described as normative competency, as an important capability for navigating complex decisions. Values-thinking involves identifying the values that influence a decision, understanding where stakeholder values differ and determining what ought to be pursued. It helps decision-makers move beyond asking, “Can we deliver this?” to consider, “Should we deliver it in this form, and for whose benefit?” For project leaders, this means being able to: 


  • recognise the value assumptions embedded in business cases and decision criteria; 

  • identify who benefits, who carries the risk and whose voice is missing; 

  • distinguish legal compliance from broader responsibility; 

  • examine short-term benefits against long-term consequences; 

  • facilitate disagreement between legitimate stakeholder perspectives; 

  • and make trade-offs transparent to accountable decision-makers. 


These capabilities are increasingly important as projects address climate resilience, digital inclusion, workforce transition, public trust and the use of AI itself. Traditional measures of cost, time and scope remain necessary. They are not sufficient to determine whether a project has created responsible and sustainable value. 


business discussion
business discussion

AI increases the need for human judgment 


The growth of AI does not reduce the need for values-thinking, it increases it. AI systems are shaped by their training data, design choices, objectives and use context. Their outputs may reflect historical patterns that organisations are trying to change. Recommendations can also appear authoritative even when important stakeholder perspectives or consequences are missing. Australia’s AI Ethics Principles emphasise human, societal and environmental wellbeing, human-centred values, fairness, transparency, contestability and accountability. These principles recognise that responsible AI requires more than technical performance. In a project environment, this means leaders must understand: 


  • where AI has influenced a recommendation; 

  • which data and assumptions shaped the output; 

  • which groups may be affected; 

  • what uncertainty remains; 

  • whether the recommendation can be challenged; 

  • and who is accountable for the final decision. 


Human oversight should not mean approving an AI-generated answer without understanding it. Nor should leaders be able to transfer responsibility to the system when consequences emerge. AI can support judgment but it cannot assume accountability. 


business plan
business plan

The limits of the project dashboard 


Dashboards are useful because they reduce complexity. They highlight selected indicators and allow leaders to see emerging patterns. But every dashboard is also a choice about what will be visible. A project may report green because it is within its approved tolerances while stakeholder trust is deteriorating. A program may appear financially healthy because the cost of operational disruption sits outside its budget. A portfolio may report benefits on track while cumulative change is overwhelming the people expected to realise them. 


The absence of a red indicator does not mean the absence of a problem. Boards and executives should therefore treat dashboards as prompts for inquiry rather than substitutes for judgment. The purpose of project information is not to make decisions automatic. It is to make better questions possible. Leaders should ask what the dashboard does not show, whose perspective is missing and whether the measures remain connected to the organisation’s purpose. 


office planning
planning

Applying PMLogic’s DELIVER framework 


PMLogic’s DELIVER framework provides a practical structure for closing the wisdom gap. 


Discover: Establish the purpose of the initiative and discover the different values, needs and expectations of affected stakeholders. Define the problem before optimising the proposed solution. 


Examine: Examine options, evidence, assumptions and consequences. Identify who benefits, who carries risk and which impacts may be missing from the analysis. 


Learn: Draw on previous experience, research, community knowledge and specialist expertise. Use AI to broaden access to evidence while recognising the limitations of historical data. 


Implement: Translate decisions into action with clear accountability. Ensure that the selected approach reflects the values and trade-offs agreed through governance, not merely the most convenient delivery option. 


Validate: Test whether the initiative is producing the intended value and whether unforeseen consequences are emerging. Validate AI-supported findings through appropriate human expertise and stakeholder engagement. 


Evaluate: Assess outcomes across financial, operational, social and environmental dimensions. Determine whether the project delivered the right change, not simply whether it delivered the approved scope. 


Reinforce: Embed lessons into governance, capability and future decision processes. Strengthen the organisation’s ability to exercise sound judgment as technology and circumstances evolve. 


DELIVER helps ensure that AI and data support responsible decisions throughout the lifecycle rather than simply producing faster analysis at the beginning. 


group project assembly
group project assembly

Questions leaders should ask at critical decisions 


Before accepting a major project recommendation, boards, sponsors and project leaders should ask: 

  1. What outcome are we optimising, and who selected it? 

  2. Which values are embedded in the decision criteria? 

  3. Who benefits from this option and who carries its costs or risks? 

  4. Which consequences are not represented in the data? 

  5. What assumptions would materially change the recommendation? 

  6. How has AI influenced the analysis? 

  7. Can affected stakeholders understand and challenge the decision? 

  8. Who remains accountable for the final judgment? 

  9. Would we consider this a good decision if its full consequences became public? 


These questions do not prevent timely decision-making. They improve the likelihood that speed will not be mistaken for wisdom. 


Closing Australia's delivery challenge 


Across this series, we have argued that Australia’s delivery challenge is not simply a shortage of projects, people, technology or information. It is a challenge of making choices. We must decide which investments should proceed, how limited capacity should be allocated, how much change organisations can absorb and where AI should support rather than substitute for human leadership. 


The final challenge is deciding well when the available evidence is extensive but the values remain contested. Better data can improve project decisions. AI can increase the speed, scale and sophistication of analysis. Neither can remove the need for leaders to examine purpose, balance competing interests and remain accountable for consequences. The future of project leadership will not be defined only by who has access to the most powerful technology. It will be defined by who has the wisdom to use it responsibly. 


professional woman portrait
professional woman portrait

How PMLogic can help 


office work meeting

PMLogic helps boards, executives and delivery leaders strengthen project governance, assurance and AI-enabled decision-making while maintaining clear human accountability. 


If you are considering how AI, sustainability and competing stakeholder values should influence critical project decisions, contact the PMLogic team for a confidential discussion about strengthening values-thinking and responsible judgment across your delivery environment. 



 

Comments


bottom of page