Strategic Governance in High-Entropy Information Environments: A Framework for Decision-Making Amidst Systemic Data Failure

 

The contemporary corporate landscape is increasingly characterized by a profound contradiction: while the volume of digital signals generated by global commerce has expanded exponentially, the reliability of the information derived from these signals—the "data"—has undergone a simultaneous process of entropy. This degradation, often colloquially termed "garbage data," represents a fundamental threat to the efficacy of corporate strategy and the stability of resource allocation models. The impact of poor data quality is far-reaching, influencing not only the immediate accuracy of analytical outputs but also the deeper cultural and structural health of the enterprise.1 As information has transitioned from a supporting asset to the primary "organizational currency," its potential for mismanagement has grown proportionally. The subsequent analysis explores the mechanisms of data decay, the financial and psychological toll of information failure, and the sophisticated frameworks required to maintain strategic momentum when traditional data-driven models collapse.

The Taxonomy of Information Decay and Data Entropy

The systemic failure of data quality within a typical enterprise is rarely the result of a single localized error; rather, it is a consequence of fragmented data ecosystems, the persistence of legacy systems, and a pervasive lack of standardized assessment practices.1 Data entropy manifests through several distinct but interrelated dimensions: accuracy, completeness, consistency, timeliness, and uniqueness.2 In the public sector and large-scale enterprises alike, these issues are exacerbated by a lack of governance and the scarcity of resources dedicated to data hygiene.1

A primary driver of this entropy is "data decay," a phenomenon most visible in marketing and sales departments where contact information, customer preferences, and lead attributes deteriorate over time. Research suggests that approximately 40% of email users change their email addresses every two years, illustrating the rapid half-life of static datasets.4 When these datasets are not continuously refreshed, the resulting "garbage" leads to missed market trends, damaged customer relations, and skewed analytical conclusions.4

Dimension of Data Quality	Description of Indicator	Strategic Risk of Failure
Accuracy	The degree to which data correctly reflects the real-world entity or event it describes.2	Flawed pipeline revenue forecasts; erroneous territory planning based on distorted customer counts.2
Completeness	The extent to which all required values are present within the dataset.2	Marketing automation failure; exclusion of viable customer segments from strategic initiatives.2
Consistency	The degree to which data is equivalent across disparate systems and silos.2	Conflict over "version of truth" during audits; revenue leakage due to mismatched discounting tiers.2
Timeliness	The currency of the information; the time elapsed since the last update.2	Rapid decay of lead relevance; strategic responses based on obsolete market conditions.4
Uniqueness	The absence of duplicate records for a single entity.3	Skewed performance metrics; wasted outreach spend; fragmented customer histories.3

The presence of near-duplicates in a database can introduce a mean absolute error (MAE) of approximately $0.100$ on performance models, representing an adverse influence of 1.1% on overall modeling effectiveness.1 While simple deduplication can yield quick efficiency gains, the resolution of near-duplicates requires a more comprehensive approach that is often only justifiable in specific business contexts.1 The distribution of data across different locations and formats further complicates this landscape, as heterogeneous data structures with varying quality levels make it difficult to align the enterprise toward a single strategic goal.1

The Economic and Cultural Toll of Poor Data Quality

The financial implications of operating with poor data quality are significant and well-documented. Organizations lose an average of $13 million to $15 million annually due to lapses in data integrity.4 These losses are not merely abstract figures but are the result of tangible operational failures: products shipped to incorrect addresses, lost customers due to erroneous client information, and damaged reputations stemming from misleading product specifications.4

The "productivity loss" associated with data failure is a critical component of these costs. Gartner has identified that thousands of dollars are lost per employee each year as staff spend excessive time manually researching and correcting errors in incomplete datasets.4 This inefficiency is particularly damaging in organizations that utilize a "data silo" philosophy, which makes the correction of centralized errors functionally impossible.4

Beyond the balance sheet, poor data quality exerts a corrosive influence on organizational culture and employee morale. Low-quality data breeds mistrust and makes it increasingly difficult to align disparate departments.1 For sales and marketing teams, the frustration of pursuing bad leads based on faulty data can destroy faith in leadership.4 This psychological toll is compounded when employees are forced into a constant state of manual reconciliation—a tedious process that diverts attention from high-value strategic tasks and contributes to disengagement and turnover.7

In sectors where data reliability is a matter of safety, such as healthcare, the cost of "garbage data" can transcend financial loss. Incorrect patient records can lead to the prescription of drugs that trigger fatal reactions, creating not only humanitarian tragedies but also catastrophic legal and reputational risks.6 Thus, the quality of data is paramount whenever it is tethered to the three core activities of any enterprise: making money, spending money, or managing risk.6

Structural Barriers to Organizational Alignment

Many organizations attempt to solve the "garbage data" problem through technological interventions, such as the adoption of new dashboard tools or the consolidation of reporting platforms. However, alignment is an enterprise-level decision rather than a technical project.7 Real alignment starts with a strategy-level conversation regarding how the business defines success, how teams measure impact, and how decisions are actually made.7

A common failure point is the lack of a "unified data foundation" between marketing and sales teams.8 While these departments often attempt to align through increased communication and shared meetings, true operational reality is only achieved through shared analytics.8 Without a common data bedrock, teams inevitably work from conflicting assumptions regarding pipeline health and revenue contribution. Marketing dashboards may reflect campaign success while sales dashboards show stalled deals; both may be accurate in isolation but disconnected from the larger strategic picture.8

This misalignment often remains invisible until it is too late, surfacing as missed opportunities, slow decision-making, and clashing priorities.7 In times of market uncertainty, the margin for error shrinks, and the friction introduced by misaligned metrics can cause leadership to overreact or underreact to performance signals.7 Furthermore, the lack of clean and consistent data limits an organization's ability to adopt advanced technologies such as Artificial Intelligence (AI) or machine learning, as these tools often reinforce bad assumptions when trained on "garbage" data.7

Navigating Uncertainty via the Cynefin Diagnostic

When data is unreliable, the first task of the executive is to diagnose the nature of the environment in which they are operating. The Cynefin framework, a decision-making model developed by David J. Snowden, categorizes organizational challenges into five distinct domains: Simple, Complicated, Complex, Chaotic, and Disorder.9 Each domain requires a fundamentally different approach to information processing and decision-making.

In the Simple (Clear) domain, cause-and-effect relationships are obvious and stable. Data is generally reliable, and the appropriate response is to "Sense – Categorize – Respond" by following established best practices or standard operating procedures.11 However, the primary danger in this domain is complacency; if the context shifts and the leader misses the change, the organization can rapidly tumble into the Chaotic domain.10

The Complicated domain is characterized by "known unknowns," where the relationship between cause and effect requires expert analysis. There may be multiple right answers, and the role of the leader is to "Sense – Analyze – Respond".12 Data in this domain is often messy or incomplete, but it can be rendered useful through technical expertise and expert intervention, such as financial audits.11

The Complex domain is the "domain of emergence," where the relationship between cause and effect can only be understood in retrospect.10 In this environment, "garbage data" is frequently the result of unpredictable system dynamics. Leaders must adopt an "experimental management" style, utilizing a "Probe – Sense – Respond" approach.10 Rather than demanding fail-safe business plans, leaders should create environments that allow patterns to emerge, using small-scale experiments and "positive deviance" to identify solutions that are already working within the organization.10

In the Chaotic domain, there is no discernible relationship between cause and effect. Immediate action is required to restore order.9 Data is often non-existent or decays so rapidly that it is useless for traditional analysis. The leader must "Act – Sense – Respond" to stabilize the situation and transition it back to a manageable state.12

Finally, the Disorder domain occurs when it is unclear which of the other four domains is predominant. The strategy here is to break the situation into constituent parts and assign each to its appropriate domain.10

Cynefin Domain	Nature of Relationship	Effective Response Strategy	Typical Decision Style
Simple / Clear	Obvious, stable cause-effect.11	Sense – Categorize – Respond.12	Procedural; Best Practice.10
Complicated	Requires expert analysis; multiple right answers.12	Sense – Analyze – Respond.12	Investigative; Expert-led.12
Complex	Emergent patterns; cause-effect clear only in hindsight.10	Probe – Sense – Respond.10	Experimental; Adaptive.10
Chaotic	No cause-effect relationship; urgent action needed.12	Act – Sense – Respond.12	Decisive; Rapid intervention.12
Disorder	Unclear which domain applies.10	Break down and categorize.10	Diagnostic; Multi-method.10

The Pivot to Decision-Driven Analytics

The traditional "data-driven" approach often fails because it focuses on the data itself rather than the decisions that need to be made. Many organizations pour resources into analytics teams only to find that the resulting insights are "floating untethered from any actual business choice".13 The alternative is "Decision-Driven Analytics," a framework that reverses the sequence of information processing.13

In this model, the organization starts by identifying the actual decision it needs to make—not a vague aspiration, but a concrete choice between real alternatives.14 Once the decision is clarified, the team asks: "What information would actually change our mind?".14 Only after defining the questions and the alternatives does the collection and analysis of data begin. This approach prevents the "divers"—those who thrive on technical depth—from becoming disconnected from the "runners"—those driven by market intuition and experience.14

A major pillar of this framework is the distinction between factual and counterfactual questions.15 Factual questions look at what is likely to happen based on existing patterns, whereas counterfactual questions examine what would happen under different choices or interventions.15 Leaders often crave precise, definitive numbers, believing that specificity conveys authority; however, in a world of "garbage data," precise answers are frequently misleading. The decision-driven approach encourages leaders to use uncertainty as a tool for uncovering hidden assumptions rather than demanding artificially crisp answers.15

Heuristic Intelligence and the "Less is More" Effect

Under conditions of extreme uncertainty and poor data integrity, the pursuit of more information can actually lead to worse decisions—a phenomenon known as the "less-is-more" effect.16 When the environment is unstable, complex statistical models often suffer from "overfitting," where they become so finely tuned to the noise of historical data that they fail to predict future outcomes.17

Fast-and-frugal heuristics—mental shortcuts or "rules of thumb"—are often more robust than complex models because they ignore less salient information, which frequently contains the most noise.17 These heuristics are "fast" because they do not require massive computational effort and "frugal" because they use minimal information.16

The Take-the-Best Heuristic and Bounded Rationality

One of the most effective heuristics for corporate decision-making is "Take the Best" (TTB). This rule involves searching through available information in order of its historically proven usefulness and stopping as soon as one piece of information favors one alternative over another.16 In experiments comparing TTB to multiple linear regression across a wide range of predictions—such as house prices, mortality rates, and dropout rates—the simpler heuristic often demonstrated superior predictive power in new environments.16

This "ecological rationality" recognizes that the human mind is adapted to making decisions with limited time and information.19 In fields like medicine and sport, where perfect information does not exist, heuristics are not "second-best" fallbacks but are the only viable option when optimization is mathematically intractable.19

Simple Rules for Complex Strategies

In the organizational domain, heuristics can be codified as "Simple Rules" that coordinate potentially large groups of people without the need for bureaucratic structure.21 These rules fall into several functional categories:

• Boundary Rules: These guide the choice of what to do and what not to do without requiring extensive analysis. For example, a venture capital firm might use a rule to "only invest in unique ideas" to filter through thousands of opportunities.22

• Prioritizing Rules: These rank alternatives competing for scarce resources, such as assigning engineers to the product with the shortest time-to-market.22

• Stopping Rules: These help organizations recognize when to "pull the plug" on a failing investment, mitigating the sunk-cost trap.22

• How-To Rules: These guide the execution of specific tasks, such as Tina Fey’s production rule for Saturday Night Live: "never tell a crazy person he's crazy".22

By capping the number of rules and tailoring them to the specific organizational context, leaders can maintain agility while ensuring that everyone remains focused on what matters most.22

Type of Simple Rule	Objective	Practical Example
Boundary Rules	Screen opportunities quickly.22	"Enter English-speaking markets first".23
Prioritizing Rules	Manage resource bottlenecks.22	Allocate advertising spend based on SKU velocity.22
Timing Rules	Orchestrate sequences and rhythms.22	Launch new products every six months to maintain market momentum.22
Stopping Rules	Mitigate sunk-cost bias.22	Halt R&D if a prototype fails to meet safety specs within 12 months.22
Coordination Rules	Align multiple actors.22	The "15% rule" at 3M allowing researchers to pursue personal projects.10

Adversarial Decision Architecture: Red Teaming and Pre-Mortems

When the data is unreliable, the executive must build "muscle memory" before a crisis occurs by stress-testing their strategy through adversarial methods.24 These techniques are designed to overcome human frailties and cognitive biases that cause faulty judgments to become "cemented" in the minds of decision-makers.25

The Pre-Mortem Protocol

The "Pre-Mortem" is a technique that uses "prospective hindsight" to identify potential failures before they happen.26 The team is instructed to imagine they are one year in the future and the strategy has failed spectacularly. Each participant must independently write down as many reasons for this failure as possible.26 Because the failure is framed as a past event, the exercise gives people "permission to dissent" and makes criticism an explicit task rather than an act of disloyalty.26

Red Teaming and the Adversarial Mandate

Red Teaming involves assigning a group—separate from the original planning team—to "attack" the strategy.26 Their mandate is to find flaws, identify vulnerabilities, and take the perspective of competitors, regulators, or skeptical investors.26 This adversarial approach is essential for identifying organizational gaps that standard data reporting might obscure.27 In high-stakes environments, leaders should use Red Teams to challenge the validity of quantification assumptions and ensure that supply-chain or inherited risks have been considered.28

Another specialized technique is "Ritual Dissent," where parallel teams critique each other's conclusions "no holds barred".10 This process hones ideas and ensures that only the most robust strategies survive the deliberation phase. These methods shift the decision-making process from one of "advocacy"—where participants selectively present data to support their preferred solution—to one of "inquiry"—where the focus is on fostering productive conflict and considering a broad range of options.29

Strategic Foresight and Scenario Design

In uncertain business systems, organizational survival depends on "strategic agility," which is enabled by strategic foresight.30 Scenario planning is a key tool in this process, allowing leaders to make sense of the unknown future by developing a series of alternatives in response to potential external shocks.30

Perceiving, Prospecting, and Probing

Effective corporate foresight involves three core practices:

1. Perceiving: Identifying drivers of change—the "signals" in the noise—to lessen uncertainty.30

2. Prospecting: Determining the potential effects of these signals on the organization.30

3. Probing: Deciding on specific actions the organization should take to navigate the emerging landscape.30

Scenarios must focus on "plausible" rather than merely "possible" futures, thereby incorporating the element of likelihood.30 This requires a willingness to invest significant time—often entire mornings or afternoons—to "unpack" systemic forces like socio-political trends and competitor behavior.30 The resulting scenarios must be internally consistent, compelling, and actionable, providing a link between "business as usual" and the "business unusual" of the future.30

Communicating Uncertainty to Boards and Investors

One of the most dangerous traps in corporate decision-making is the "lure of incredible certitude," where researchers and managers report point predictions despite high levels of scientific or market uncertainty.32 When executives fail to adequately explain the limitations of the data underlying their judgments, it creates a "serious weakness" in the governance chain.33

The communication of uncertainty should be structured around Lasswell’s model: who communicates what, in what form, to whom, and to what effect.33 Leaders must be transparent about the object of uncertainty, the source of the lack of knowledge, and the magnitude of the ignorance.33 Research suggests that communicating uncertainty numerically (e.g., using probability distributions or ranges) is often perceived as more reliable than using purely verbal descriptors, which can be interpreted differently by different stakeholders.33

Board Governance and the Litmus Test of Reporting

Effective data governance at the board level is foundational to protecting the organization's strategic assets and meeting regulatory requirements.5 Boards should promote an "analytics mindset" that encourages data-informed decisions while acknowledging associated risks.5

A critical "litmus test" for board effectiveness is the quality of the board reporting itself. Directors should use the clarity and timeliness of the data in their own "board packs" as an indicator of how well data is being governed throughout the rest of the organization.5 If the board pack is filled with technical jargon or fails to demonstrate the accuracy of its metrics, it is a "red flag" that data silos and quality issues are likely endemic within the operational layers of the firm.5

Governance Mechanism	Board Responsibility	Strategic Objective
Monitoring Quality Metrics	Review indicators of accuracy, completeness, and timeliness.5	Verify reliability of data informing strategic choices.5
Risk Oversight	Understand data lifecycle controls and external provider handling.5	Prevent failure in major strategic initiatives due to "garbage" inputs.5
ROI Evaluation	Measure data investments in terms of value creation and cost savings.5	Ensure data products power the "flywheel effect" of accelerating value.5
Scenario Oversight	Utilize scenario planning to manage strategy during uncertainty.31	Develop resilient and forward-looking contingency plans.31

The Role of Strategic Judgment and "Altitude"

Ultimately, when the data is "garbage," the most valuable asset an organization possesses is the strategic judgment of its leaders. This judgment involves the ability to "rise above the noise" to see where the business is truly headed—a quality known as "altitude".35 Leaders must interpret dashboard numbers by considering the broader business context, including vendor ecosystems, regulatory exposure, and human factors.28

Successful risk-takers are process-oriented rather than results-oriented; they focus on the "long game" while being comfortable with the crude estimates required to act in high-velocity environments.35 They recognize that while data is the "fuel" for the modern marketing and sales engine, nearly half of that fuel is often "contaminated".3 Resilience stems not from having perfect data, but from having a "narrative understanding" of the world and a governance structure that can withstand the inevitable failures of information.28

Strategic Implications for the Future

As companies increasingly rely on data to innovate—from digital twins to generative AI—the challenge of "garbage data" will only intensify. Ineffective data product practices have already become a top strategic issue.34 To scale successfully, organizations must focus on "value" rather than just "better data," ensuring that every data product is built for reuse and aligns with meaningful KPIs such as time-to-value and stakeholder trust.34

In conclusion, making corporate decisions when the data is garbage requires a multi-dimensional strategy. It necessitates a diagnostic understanding of the environmental context via Cynefin, a shift to decision-driven analytics, the adoption of robust heuristics for bounded rationality, and the institutionalization of adversarial stress-testing. By embracing uncertainty as a realistic reflection of the market and focusing on the "big decisions" that drive the majority of outcomes, the enterprise can maintain strategic clarity even in the most high-entropy information environments. The goal is not to find a purpose for the data at hand, but to find the right information for a clearly defined purpose, thereby transforming "garbage" into a catalyst for strategic discipline and organizational resilience.

Works cited

1. The Impact of Poor Data Quality on the Typical Enterprise. - ResearchGate, accessed January 20, 2026, https://www.researchgate.net/publication/27295794_The_Impact_of_Poor_Data_Quality_on_the_Typical_Enterprise

2. How Inaccurate Data Impacts Your Bottom Line - Data Ladder, accessed January 20, 2026, https://dataladder.com/how-inaccurate-data-impacts-your-bottom-line/

3. Almost 50% of Marketing Data is Inaccurate, Reveals New Research - Adverity, accessed January 20, 2026, https://www.adverity.com/press-releases/almost-50-of-marketing-data-is-inaccurate-reveals-new-research

4. The Impact of Poor Data Quality (and How to Fix It) - Dataversity, accessed January 20, 2026, https://www.dataversity.net/articles/the-impact-of-poor-data-quality-and-how-to-fix-it/

5. Data Governance Foundations for Boards - Australian Institute of ..., accessed January 20, 2026, https://www.aicd.com.au/content/dam/aicd/pdf/tools-resources/director-tools/organisation/data-governance-foundations-for-boards-web.pdf

6. A Framework to Understand How Poor Data Quality Hurts Business Performance, accessed January 20, 2026, https://www.metaplane.dev/blog/how-poor-data-quality-hurts-business-performance

7. What misaligned data is really costing you | Wipfli, accessed January 20, 2026, https://www.wipfli.com/insights/articles/what-misaligned-data-is-really-costing-you

8. Why B2B Marketing and Sales Alignment Fails Without Shared Data - DemandZEN, accessed January 20, 2026, https://demandzen.com/b2b-marketing-and-sales-alignment-shared-data/

9. accessed January 20, 2026, https://creately.com/guides/understanding-the-cynefin-framework/#:~:text=The%20Cynefin%20Framework%20divides%20decision,for%20decision%2Dmaking%20and%20leadership.

10. A Leader's Framework for Decision Making - Center for Homeland ..., accessed January 20, 2026, https://www.chds.us/ed/resources/uploads/2020/10/A-Leaders-Framework-for-Decision-Making-Snowden-and-Boone.pdf

11. Choosing the Right Decision-Making Framework: A Comprehensive Guide for Organizational Success - Service Quality Centre, accessed January 20, 2026, https://www.sqcentre.com/blog/choosing-the-right-decision-making-framework-a-comprehensive-guide-for-organizational-success/

12. What is the Cynefin Framework, and How to Use it? - Creately, accessed January 20, 2026, https://creately.com/guides/understanding-the-cynefin-framework/

13. Decision-Driven Analytics - Penn Press, accessed January 20, 2026, https://www.pennpress.org/9781613631713/decision-driven-analytics/

14. Decision-Driven Analytics Summary of Key Ideas and Review | Bart ..., accessed January 20, 2026, https://www.blinkist.com/en/books/decision-driven-analytics-en

15. Decision-Driven Analytics: Leveraging Human Intelligence to Unlock the Power of Data, accessed January 20, 2026, https://www.goodreads.com/book/show/202559506-decision-driven-analytics

16. Bounded Rationality: the Case of 'Fast and Frugal' Heuristics - Exploring Economics, accessed January 20, 2026, https://www.exploring-economics.org/en/discover/bounded-rationality-heuristics/

17. Gigerenzer - Fast and Frugal Heuristics - The Broken Science Initiative, accessed January 20, 2026, https://brokenscience.org/gigerenzer-heuristic/

18. Full article: Fast-and-frugal heuristics: an exploration into building an adaptive toolbox to assess the uncertainty of rework, accessed January 20, 2026, https://www.tandfonline.com/doi/full/10.1080/09537287.2023.2257178

19. Fast-and-frugal heuristics: analytical models of intuition | IMA Journal of Management Mathematics | Oxford Academic, accessed January 20, 2026, https://academic.oup.com/imaman/advance-article/doi/10.1093/imaman/dpaf041/8293995

20. The power of simplicity: a fast-and-frugal heuristics approach to performance science - Frontiers, accessed January 20, 2026, https://www.frontiersin.org/journals/psychology/articles/10.3389/fpsyg.2015.01672/full

21. Simple Rules and Other Heuristics in Strategy and Organizational Research, accessed January 20, 2026, https://oxfordre.com/business/display/10.1093/acrefore/9780190224851.001.0001/acrefore-9780190224851-e-458?d=%2F10.1093%2Facrefore%2F9780190224851.001.0001%2Facrefore-9780190224851-e-458&p=emailAUGEae3nzCzAQ

22. Simple Rules: How to Thrive in a Complex World - Farnam Street, accessed January 20, 2026, https://fs.blog/simple-rules/

23. The power of simpler strategy | UNC Kenan-Flagler, accessed January 20, 2026, https://www.kenan-flagler.unc.edu/news/the-power-of-simpler-strategy/

24. How Stressful Is the CEO Role—& How Great Leaders Cope [12 Key Factors][2026], accessed January 20, 2026, https://digitaldefynd.com/IQ/is-being-ceo-a-stressful-job/

25. Red Teaming Handbook, 3rd Edition - GOV.UK, accessed January 20, 2026, https://assets.publishing.service.gov.uk/media/61702155e90e07197867eb93/20210625-Red_Teaming_Handbook.pdf

26. Chapter 4: Developing Strategic Intuition - The Strategy Engine: Business Models, Unit Economics & Moats, accessed January 20, 2026, https://strategy-engine.pages.dev/chapters/chapter-04-strategic-intuition/

27. Red-Teaming in the Public Interest | Data & Society, accessed January 20, 2026, https://datasociety.net/wp-content/uploads/2025/02/Red-Teaming-in_the_Public_Interest_FINAL1.pdf

28. Cyber risk in the boardroom: Why judgment matters more than numbers - Forbes India, accessed January 20, 2026, https://www.forbesindia.com/article/thought-leadership/iim-calcutta/cyber-risk-in-the-boardroom-why-judgment-matters-more-than-numbers/2990483/1

29. (PDF) What You Don't Know about Making Decisions - ResearchGate, accessed January 20, 2026, https://www.researchgate.net/publication/11796182_What_You_Don't_Know_about_Making_Decisions

30. Scenario-planning in strategic decision-making: requirements ..., accessed January 20, 2026, https://repository.up.ac.za/bitstreams/cf831575-b726-4b04-9f77-f6c80f5dd57a/download

31. Six Steps for Boards to Manage Strategy During Times of Uncertainty, accessed January 20, 2026, https://mpti.com.au/six-steps-for-boards-to-manage-strategy-during-times-of-uncertainty/

32. The lure of incredible certitude | Economics & Philosophy | Cambridge Core, accessed January 20, 2026, https://www.cambridge.org/core/journals/economics-and-philosophy/article/lure-of-incredible-certitude/A1F09783377B05F20B83543CD40C7639

33. (PDF) 18 The effects of communicating uncertainty about facts and numbers - ResearchGate, accessed January 20, 2026, https://www.researchgate.net/publication/328460369_18_The_effects_of_communicating_uncertainty_about_facts_and_numbers

34. The missing data link: Five practical lessons to scale your data products - McKinsey, accessed January 20, 2026, https://www.mckinsey.com/capabilities/tech-and-ai/our-insights/the-missing-data-link-five-practical-lessons-to-scale-your-data-products

35. The Leading Blog - Leadership Now, accessed January 20, 2026, https://www.leadershipnow.com/leadingblog/problem_solving/