The Measurement Layer Beneath Every Capital Program

By Eric Knauf

The Capital Allocation

Every major capital program a board approves rests on a workforce execution assumption no one has measured. A merger and acquisition integration is priced on a synergy case that assumes the combined workforce, at the manager-unit level, can execute redesigned operating models, integrated systems, and consolidated functions. A private equity operating thesis assumes the workforce inherited at close can carry the value-creation plan, the hundred-day plan, and the operating model redesign that follows.

A large-scale transformation assumes the workforce can absorb the redesign and produce the modeled return. An AI transformation assumes the workforce can adopt the tools, redesign the work, and capture the value released by automation.

These are not assumptions about technology, process, or finance. They are assumptions about human infrastructure, the behavioral and organizational properties at the manager-unit level that decide whether a workforce can do what the model says it will do. In The 56% Solution, I describe this as the difference between programs and infrastructure: programs depend on individual champions, isolated initiatives, and good intentions, whereas infrastructure is the set of conditions and systems that determine whether other initiatives will succeed (Knauf, 2025).

No widely adopted instrument measures whether those conditions hold at the layer where execution happens. Engagement surveys measure sentiment on the day they are administered. Climate surveys measure perception of recent events.

Diversity dashboards measure representation. Organizational health tools operate at enterprise aggregates and obscure the manager-unit variance that determines whether the program is executable.

AI makes the gap most visible. A 2025 Massachusetts Institute of Technology report of enterprise generative AI initiatives found that, in its sample, roughly 5 percent of pilots delivered measurable multi-million-dollar value while the remainder produced little or no P&L impact, with the authors identifying organizational learning gaps in feedback loops and workflow integration as a primary barrier (MIT Project NANDA, 2025). Large studies of merger integration, transformation, and value-creation execution attribute unrealized value to the same organizational and people factors. The category is not AI. The category is the measurement layer beneath every capital program, where the workforce execution assumption sits unmeasured.

Belonging infrastructure is that measurement layer. Organized across five jointly necessary pillars and measured at the manager-unit level, it is the missing instrument that determines whether the workforce can execute what the capital allocator approved.

The article that follows makes the case across seven sections:

1. Human infrastructure and identifies belonging infrastructure as the measurable layer beneath it; 
2. Why existing instruments fail to measure that layer; 
3. The same gap appears across all four classes of capital programs. 
4. Specifies what a board-grade instrument must do to close the gap; 
5. Evidence from one engagement clarifies that the instrument itself is next-generation work in pilot. 
6. The governance trajectory that is moving boards toward instrument-grade oversight of human capital, 
7. An outline of what changes for boards that build the measurement capability now, and what continues for boards that do not.

What Human Infrastructure Is

Human infrastructure is the set of behavioral and organizational properties at the manager-unit level that determine whether a workforce can execute a given capital program. The conditions are not feelings. They are observable, measurable, and variable across teams within the same enterprise.

Belonging infrastructure is the measurable layer beneath human infrastructure. It is organized across five pillars: Psychological Safety, Inclusion, Support, Connection, and Purpose. Belonging is a fundamental human motivation and a distinct, measurable construct (Baumeister & Leary, 1995; Hagerty & Patusky, 1995), and industry studies show that workplace belonging predicts engagement, retention, and performance across industries (Achievers Workforce Institute, 2021; BetterUp, 2019; Berkeley Othering & Belonging Institute, 2023). In The 56% Solution, I define each pillar with observable behaviors and link each to the operational outcomes it produces (Knauf, 2025).

Psychological Safety is the foundation. People speak up, share concerns, admit mistakes, and challenge ideas without fear of interpersonal risk or career penalty. Edmondson's research on team learning and Clark's sequential model of safety stages establish the empirical basis (Edmondson, 1999; Clark, 2020).

Inclusion means diverse perspectives shape decisions and outcomes, not as invited participants but as participants whose input changes the result. Research on inclusive climates shows that when people experience an inclusive climate, diversity yields higher performance and lower turnover; when inclusion is weak, representation alone yields neither (Nishii, 2013; McKinsey & Company, 2020).

Support means resources, tools, guidance, and advocacy that reach people before they break down or burn out. The Job Demands-Resources literature documents the mechanism (Demerouti et al., 2001; Bakker and Demerouti, 2017).

Connection refers to trust-based relationships that extend beyond immediate job requirements. High-quality connections at work create energy, vitality, and capacity for individuals and organizations (Dutton & Heaphy, 2003), and integrative models of organizational trust treat ability, benevolence, and integrity as the necessary components for the trust that makes collaboration work (Mayer, Davis, & Schoorman, 1995). Zak's research on trust and collaboration documents the link to coordination effectiveness (Zak, 2017).

Purpose is the understanding of how individual work contributes to meaningful outcomes, combined with recognition that reinforces the contribution. Research on meaningful work shows that identity connection and purpose alignment drive engagement and belonging (Pratt & Ashforth, 2003; Rosso, Dekas, & Wrzesniewski, 2010), and purpose-driven work produces performance advantages (Hurst, 2014).

The five pillars are jointly necessary. BelongHQ internal validation studies use structural equation modeling across diverse organizational samples to examine how the pillars interact (Knauf, 2025). Uneven pillar development produces lower overall performance than balanced moderate scores, and single-pillar improvements diminish when the other conditions lag. The pattern aligns with broader Job Demands-Resources research on how multiple resources together buffer demands and enable engagement (Bakker & Demerouti, 2007; Christian, Garza, & Slaughter, 2011).

The framework measures all five pillars on the same instrument, on the same schedule, across the same population. A subset is not the framework. The instrument scores each pillar on a one-to-five scale at the manager-unit level. The composite is proprietary and not published, and the unit of analysis is the manager and the team because the variance that predicts program execution sits below the enterprise aggregate.

Diagram 1: Belonging infrastructure is the measurable layer beneath every capital program: the foundation of psychological safety, the infrastructure of inclusion, support, and connection, and the activation layer of purpose that determines whether the workforce can execute what the capital allocator has approved.

Why Existing Tools Miss the Layer

Engagement surveys are designed to measure how employees feel on the day they answer. They are not designed to measure the conditions that determine whether a workforce can execute a capital program. Sentiment is not capacity, and a high engagement score can coexist with a failed transformation within the same enterprise.

Climate surveys measure perception. Perception responds to recent events, communication, and management visibility, and it can lag or lead the conditions that drive performance, depending on the cycle. Climate is informative, but it does not isolate the organizational properties that survive leadership changes and budget pressure.

Diversity dashboards measure representation. Representation is necessary but not sufficient. Studies find that representation alone does not produce the conditions under which diverse perspectives shape decisions (McKinsey & Company, 2020). Representation without inclusion is one of the patterns that make existing tools insufficient for capital-program governance.

Organizational health and culture tools operate at enterprise aggregates, and the variance that predicts execution is not at the aggregate. Research on managerial quality and team climate finds that within-firm variance across teams is as large as or larger than the variance across firms on the dimensions that predict performance. Google's Project Aristotle, building on Edmondson's team-learning literature, made this finding visible at scale (Duhigg, 2016).

An enterprise belonging score has the same diagnostic value as an enterprise average headcount number. It describes a population. It tells you nothing about the distribution.

Existing tools were built for other purposes. They were not built to resolve manager-unit variance, link that variance to specific capital programs, and present the result on a board dashboard. The instrument that capital-program governance requires sits at a different layer.

Four Capital Programs

The same five pillars and the same manager-unit measurement apply across four classes of capital programs. The measurement gap is the same in each. The standard instruments in each domain measure sentiment, adoption activity, or representation, not the organizational properties at the manager-unit level that the program depends on.

Merger and acquisition integration. The deal model prices the synergy case, and the synergy case prices the combined workforce executing redesigned operating models, integrated systems, and consolidated functions. Longitudinal studies of post-close performance attribute a substantial share of unrealized synergy to organizational and people factors rather than to financial or operational design (Marks and Mirvis, 2015). The integration plan does not include an instrument to assess whether the conditions for execution are met at the manager-unit level in either the acquirer or the target. Diligence reads financial statements, talent reviews, and engagement scores, but none of those measures the organizational properties on which the synergy case rests.

Private equity value-creation execution. The operating thesis prices the workforce inherited at close. The hundred-day plan, the value-creation plan, and the operating model redesign assume execution capacity that no diligence instrument measures. Pre-close diligence misses execution drag and management gaps, and operating partners discover them after close. Portfolio-level engagement scores, NPS, and culture-fit assessments are sentiment outputs that do not measure the manager-unit conditions on which the operating thesis depends.

Large-scale transformation programs. Enterprise resource planning implementations, shared services consolidations, technology modernizations, and operating model redesigns are governed as technology and process programs. McKinsey's longitudinal transformation research stream has documented that most major transformation programs fail to meet their original objectives, with human and organizational factors identified as the primary drivers of success or failure (McKinsey, 2025). The change management workstream substitutes activity for measurement, and pulse surveys substitute sentiment for capacity.

Artificial intelligence transformation. The freed-capacity case prices a workforce that can redesign work, adopt the tools, and capture the value released by automation. The MIT GenAI Divide report fids that in its sample, roughly 95 percent of enterprise pilots produced little or no P&L impact, with organizational learning gaps identified as the primary barrier (MIT Project NANDA, 2025).

McKinsey & Company 's State of AI Global Survey 2025 reports that many organizations remain in pilots and a minority have scaled AI to deliver measurable impact, while Boston Consulting Group (BCG) 's January 2026 workforce transformation research attributes 70 percent of AI value to people and processes (McKinsey, 2025; BCG, 2026). The Conference Board 's practitioner surveys and event insights report that most firms have not yet moved beyond early AI adoption and cite organizational and change factors as the primary obstacles (Conference Board, 2025). The technology investment is governed; the human infrastructure that determines whether the technology produces the modeled return is not.

The framework in The 56% Solution was not designed for AI transformation. The five pillars govern whether teams can execute under merger stress, transformation load, value-creation pressure, or AI-driven redesign. AI is the domain where the deficits are most visible now, but the instrument applies wherever capital is allocated with a workforce-execution assumption.

What the Instrument Has to Do

For an instrument to be decision-useful at the board level on capital-program execution risk, it must satisfy three requirements.

1. Measurement at the manager-unit level, not at the enterprise aggregate level. Hierarchical linear methodology, developed by Raudenbush and Bryk for nested data structures in education and organizational research, is the standard technical implementation when the variance of interest sits across nested groups within larger populations (Raudenbush and Bryk, 2002). The variance that predicts capital-program execution is at the manager and team level. An instrument that reports enterprise averages alone cannot discharge oversight responsibility for execution risk in environments where execution capacity varies materially across manager units.

1. Simultaneous measurement of all five pillars on the same schedule, across the same population, on the same scale. Single-pillar improvements plateau without supporting infrastructure, and two-pillar or three-pillar measurement yields directional information that does not carry over to the capital-program level. The framework treats the pillars as jointly necessary, and the measurement reflects that design.

1. Board-grade output. The output identifies the capital program at risk, isolates the pillar conditions that put it at risk, and points to the management action that changes the trajectory. Sentiment outputs do not discharge oversight duties, and aggregate culture scores do not inform capital allocation decisions.

Outputs designed for human resources audiences do not survive board review. The instrument has to produce pillar-level scores at the manager-unit level, linked to specific capital programs and to the operational metrics that boards track.

Capital program outcomes are multi-factorial. Technology fit, governance design, market conditions, and execution capacity all influence whether a program produces its modeled return. The argument here is that the human infrastructure layer is one necessary input to program execution, and the input that no widely adopted instrument currently measures at board-grade fidelity.

In The 56% Solution, I describe the Belonging Standard as the measurement framework, the Maturity Model as the sequencing logic, the Audit as the gap identification mechanism, and the Platform as the system that feeds live signal data back into the program. The instrument is the application of that framework to capital-program governance. It is what the framework looks like when built to discharge the oversight duties that capital programs now require.

Practitioner Evidence

The five-pillar framework was developed and applied operationally across more than two decades of senior talent leadership in venture-backed, private equity-backed, and Fortune 500 environments. The most documented case occurred at a labor-marketplace company during a 55 percent reduction in workforce.

Employee Net Promoter Score at the start of the reduction was negative 73. Six months later, eNPS was positive 8. The shift held over the next 12 months, and voluntary attrition among the remaining workforce remained below industry benchmarks over the same period (Knauf, 2025).

The intervention was the operational application of the five conditions during and after the reduction. Psychological Safety was preserved through a two-day workshop that surfaced grievances without filter and a company-wide Voice of the Employee process that produced 76 concrete recommendations. Inclusion operated through cross-functional focus groups that returned proposals, and leadership acted on them.

Support was elevated through transition resources and proactive communication. Connection was preserved through team-level rituals during the reduction. Purpose was reaffirmed through transparent communication about the path forward and progress against the recommendations.

The eNPS movement is practitioner evidence that the five conditions, applied operationally during the most disruptive capital program a workforce can experience, produce measurable operational outcomes that hold beyond the intervention window. The current board-grade instrument was not used at that engagement; the diagnostic work was done with the underlying framework.

The instrument that operationalizes those conditions as a board-grade diagnostic for any capital program is next-generation work that builds on the framework documented in the book and on the operational pattern documented in the engagement.

The distinction matters because it sets the standard the instrument is built to meet. The framework produced the eNPS outcome at one engagement under one set of conditions. The instrument has to produce decision-useful manager-unit measurement for any capital program at any organization that funds the diagnostic, and that is the work in pilot now.

The Governance Foreshadow

The measurement gap produces a governance gap. Boards approving capital programs above material thresholds now oversee human-capital risk with the same rigor they apply to financial, technological, and operational risk.

Diagram 2: The board is approving capital against the 30 percent of AI value that is in the budget, while the 70 percent that generates value remains uninstrumented.

The Delaware courts continue to develop the Caremark line of cases, including the 2023 In re McDonald's Corporation Stockholder Derivative Litigation decision that clarified officer-level oversight duties alongside director-level duties. The Securities and Exchange Commission's Reg S-K Item 101(c) human capital disclosure requirements direct registrants to disclose material measures and objectives. Most disclosures rely on aggregate metrics, narrative descriptions, and program activity descriptions that do not connect to capital-program execution.

Boards reviewing nine-figure capital programs without an instrument that measures the human infrastructure on which those programs depend are accepting unquantified exposure in the category that published research across M&A, large-scale transformation, and AI adoption identifies as the dominant determinant of program success. The position is difficult to defend as governance expectations for human-capital oversight continue to evolve. The trajectory of disclosure standards in adjacent categories, including non-GAAP financial measures and cybersecurity governance, suggests that market practice precedes formal rules and that early instrumentation yields a governance advantage.

The next piece in this sequence addresses the governance argument: how the Caremark line and Item 101(c) intersect with human infrastructure measurement, and why instrument-grade measurement discharges those duties better than aggregate sentiment metrics.

Close

Every major capital program a board approves rests on an assumption about workforce execution. That assumption can now be measured. Belonging infrastructure, organized into five jointly necessary pillars and measured at the manager-unit level, is the instrument used to measure it.

The argument moved in a clean line. Belonging infrastructure rests on five conditions, anchored in established research and validated through BelongHQ, Inc. 's own work, which together constitute its joint necessity. Existing instruments do not see those conditions; engagement surveys are built for sentiment, climate surveys for perception, diversity dashboards for representation, and organizational health tools for enterprise aggregates that obscure the variance that decides whether capital programs execute.

The same measurement gap appears across every class of capital program, from merger integration through value-creation execution, large-scale transformation, and AI transformation. 

The instrument that closes the gap is specified: 

• measurement at the manager-unit level; 
• all five pillars on the same schedule, population, and scale; 
• output that identifies the capital program at risk and the management action that changes its trajectory. 

Operational application of the five conditions during a 55 percent workforce reduction produced eNPS movement from negative 73 to positive 8, sustained over twelve months, and the board-grade instrument that operationalizes the framework for any capital program at any organization is in pilot now. The governance trajectory is moving in the same direction, from the Caremark line through Reg S-K Item 101(c) and the precedent set in adjacent disclosure categories.

What becomes possible when the measurement layer exists changes how capital is governed. The capital case includes a measured workforce assumption alongside the financial assumption, so the board can see what the program depends on rather than approve it on faith. Caremark and Item 101(c) ask boards to oversee material risks, and instrument-grade measurement of human infrastructure makes those duties dischargeable rather than aspirational. Human capital stops being the line item that nobody measures rigorously and becomes the line item with the same governance posture as financial, technological, and operational risk.

The Belonging Standard was designed as the missing measurement layer beneath capital programs, and the instrument that operationalizes it for board-grade governance is now in pilot (Knauf, 2025). Boards that build the measurement capability will govern the next decade of capital allocation with instrumentation the previous decade never had, and the workforce execution assumption that sits beneath every M&A integration, every operating thesis, every transformation, and every AI program will be visible to the people approving the capital. Boards that do not build it will continue to underwrite execution on assumptions no one has tested, in a regulatory environment that is moving toward instrument-grade oversight, whether they prepare for it or not.

References

Achievers Workforce Institute. (2021). Belonging at work report. Achievers.

Bakker, A. B., & Demerouti, E. (2007). The job demands-resources model: State of the art. Journal of Managerial Psychology, 22(3), 309-328. https://doi.org/10.1108/02683940710733115

Bakker, A. B., & Demerouti, E. (2017). Job demands-resources theory: Taking stock and looking forward. Journal of Occupational Health Psychology, 22(3), 273-285. https://doi.org/10.1037/ocp0000056

Baumeister, R. F., & Leary, M. R. (1995). The need to belong: Desire for interpersonal attachments as a fundamental human motivation. Psychological Bulletin, 117(3), 497-529. https://doi.org/10.1037/0033-2909.117.3.497

Berkeley Othering & Belonging Institute. (2023). Places of belonging. University of California, Berkeley. https://belonging.berkeley.edu/places-of-belonging

BetterUp. (2019). The value of belonging at work: The business case for investing in workplace inclusion. BetterUp. https://www.betterup.com/press/betterups-new-industry-leading-research-shows-companies-that-fail-at-belonging-lose-tens-of-millions-in-revenue

Christian, M. S., Garza, A. S., & Slaughter, J. E. (2011). Work engagement: A quantitative review and test of its relations with task and contextual performance. Personnel Psychology, 64(1), 89-136. https://doi.org/10.1111/j.1744-6570.2010.01203.x

Clark, T. R. (2020). The 4 Stages of Psychological Safety: Defining the Path to Inclusion and Innovation. Berrett-Koehler Publishers.

Demerouti, E., Bakker, A. B., Nachreiner, F., & Schaufeli, W. B. (2001). The job demands-resources model of burnout. Journal of Applied Psychology, 86(3), 499-512. https://doi.org/10.1037/0021-9010.86.3.499

Duhigg, C. (2016, February 25). What Google learned from its quest to build the perfect team. The New York Times Magazine. https://www.nytimes.com/2016/02/28/magazine/what-google-learned-from-its-quest-to-build-the-perfect-team.html

Dutton, J. E., & Heaphy, E. D. (2003). The power of high-quality connections. In K. S. Cameron, J. E. Dutton, & R. E. Quinn (Eds.), Positive Organizational Scholarship: Foundations of a New Discipline (pp. 263-278). Berrett-Koehler.

Edmondson, A. C. (1999). Psychological safety and learning behavior in work teams. Administrative Science Quarterly, 44(2), 350-383. https://doi.org/10.2307/2666999

Edmondson, A. C., & Lei, Z. (2014). Psychological safety: The history, renaissance, and future of an interpersonal construct. Annual Review of Organizational Psychology and Organizational Behavior, 1, 23-43. https://doi.org/10.1146/annurev-orgpsych-031413-091305

The Conference Board. (2025). AI Leadership Summit insights and C-Suite Outlook practitioner surveys (2024-2026). The Conference Board.

Hagerty, B. M., & Patusky, K. (1995). Developing a measure of sense of belonging. Nursing Research, 44(1), 9-13. https://doi.org/10.1097/00006199-199501000-00003

Hurst, A. (2014). The Purpose Economy: How Your Desire for Impact, Personal Growth and Community Is Changing the World. Elevate Publishing.

In re McDonald's Corporation Stockholder Derivative Litigation, C.A. No. 2021-0324-JTL (Del. Ch. Feb. 28, 2023).

Knauf, E. (2025). The 56% Solution: How Belonging Infrastructure Transforms Performance. BelongHQ.

Marks, M. L., & Mirvis, P. H. (2015). Joining Forces: Making One Plus One Equal Three in Mergers, Acquisitions, and Alliances (3rd ed.). Jossey-Bass.

Mayer, R. C., Davis, J. H., & Schoorman, F. D. (1995). An integrative model of organizational trust. Academy of Management Review, 20(3), 709-734. https://doi.org/10.5465/amr.1995.9508080335

McKinsey & Company. (2020). Diversity wins: How inclusion matters. https://www.mckinsey.com/featured-insights/diversity-and-inclusion/diversity-wins-how-inclusion-matters

McKinsey & Company. (2025). The state of AI: Global survey 2025. QuantumBlack, AI by McKinsey. https://www.mckinsey.com/capabilities/quantumblack/our-insights/the-state-of-ai

McKinsey & Company. (2023). The economic potential of generative AI: The next productivity frontier. https://www.mckinsey.com/capabilities/mckinsey-digital/our-insights/the-economic-potential-of-generative-ai-the-next-productivity-frontier

BCG. (2026). AI transformation is a workforce transformation. Boston Consulting Group. https://www.bcg.com/publications/2026/ai-transformation-is-a-workforce-transformation

MIT Project NANDA. (2025). The GenAI Divide: State of AI in Business 2025. Massachusetts Institute of Technology.

Nishii, L. H. (2013). The benefits of climate for inclusion for gender-diverse groups. Academy of Management Journal, 56(6), 1754-1774. https://doi.org/10.5465/amj.2009.0823

Pratt, M. G., & Ashforth, B. E. (2003). Fostering meaningfulness in working and at work. In K. S. Cameron, J. E. Dutton, & R. E. Quinn (Eds.), Positive Organizational Scholarship: Foundations of a New Discipline (pp. 309-327). Berrett-Koehler.

Raudenbush, S. W., & Bryk, A. S. (2002). Hierarchical Linear Models: Applications and Data Analysis Methods (2nd ed.). Sage Publications.

Rosso, B. D., Dekas, K. H., & Wrzesniewski, A. (2010). On the meaning of work: A theoretical integration and review. Research in Organizational Behavior, 30, 91-127. https://doi.org/10.1016/j.riob.2010.09.001

Securities and Exchange Commission. (2020). Modernization of Regulation S-K Items 101, 103, and 105 (Release Nos. 33-10825; 34-89670). https://www.sec.gov/rules/final/2020/33-10825.pdf

Zak, P. J. (2017). Trust Factor: The Science of Creating High-Performance Companies. AMACOM.