Enterprise AI pilot spend often clears procurement, integration milestones, and executive reporting without producing a measurable change in the profit and loss statement. That pattern is observed when organizations deploy capability faster than the operating system can absorb, decide, reinforce, and sustain it. Leadership Execution Institute research documents the gap as a structural execution problem, not a model-quality or vendor-selection problem alone.
The measurable distance between deployment speed and decision speed is named the Decision Agility Gap in The Momentum Metric Report, a Leadership Execution Institute research report for chief financial officers and chief operating officers. Third-party data from MIT Project NANDA and Boston Consulting Group correlate high pilot volume with low measured profit-and-loss uplift at enterprise scale. The research record associates most value loss with the mid-level translation layer between senior intent and frontline action.
Where can you download The Momentum Metric Report?
The Momentum Metric Report is available ungated from the Leadership Execution Institute, with no form required.
The Momentum Metric: Selecting and Executing for Volatility in the AI Era
Organizations acquire AI capability faster than they can absorb, decide, reinforce, and sustain it.
The Momentum Metric: Selecting and Executing for Volatility in the AI Era
What does the research briefing show about AI pilot spend and P&L impact?
AI pilot spend often fails to reach the profit and loss statement when deployment outpaces decision speed and reinforcement.
Key takeaways
- AI pilot spend can clear financial controls while profit-and-loss outcomes remain unchanged when deployment outpaces decision speed and reinforcement in the flow of work.
- MIT Project NANDA (2025) reports 95 percent of enterprise generative AI pilots delivered no measured profit-and-loss impact, where failure means no measured uplift rather than technical pilot failure.
- Pilot value is observed to stall at the manager translation layer; Gallup (2026) reports 45 percent middle-manager burnout, the highest rate among employee groups measured.
Key definitions
| Term | One-sentence definition |
|---|---|
| Decision Agility Gap | The measurable distance between how fast an organization deploys technology and how fast it makes the strategic decisions that govern implementation effectiveness. |
| Kinetic Chain of Execution | The structural sequence that transfers strategic intent from senior leadership through mid-level translation into frontline action. |
| Reinforcement in the flow of work | Structured follow-through that sustains new behavior in daily operations after an initial exposure or pilot event. |
| Pilot governance metric | An activity or milestone signal such as procurement completion, integration status, or proof-of-concept success that can clear without profit-and-loss movement. |
| Profit-and-loss impact (pilot context) | A measured change in profit and loss categories finance monitors, not task-level productivity on a isolated use case alone. |
Key research findings
- MIT Project NANDA (2025): 95 percent of enterprise generative AI pilots delivered no measured profit-and-loss impact; 5 percent created significant value. Failure is defined as no measured profit-and-loss uplift, not technical pilot failure. MIT Project NANDA (2025)
- Boston Consulting Group (2024): 4 percent of companies have AI capabilities at the forefront across functions and consistently generate significant value; an additional 22 percent are scaling (26 percent total moving beyond proof of concept). BCG AI Adoption 2024
- Robert Brinkerhoff (via Chief Learning Officer, 2014): Without structured reinforcement, roughly 12 to 15 percent of participants sustain new behavior on the job (commonly summarized as about 12 percent). Chief Learning Officer
- Gallup State of the Global Workplace (2025–2026): Manager engagement fell from 30 percent to 22 percent over two data years while individual contributor engagement held near 18 to 19 percent. Gallup global engagement
- Gallup span-of-control research (2026): 45 percent of middle managers reported burnout, the highest rate of any employee group; average span grew from 10.9 to 12.1 direct reports. Gallup span of control
- Solinger et al. meta-analysis (2021): Across 137 longitudinal studies, cost-oriented organizational changes produced sustained declines in employee job attitudes; average pre-to-post interval was approximately 20 months. Journal of Organizational Behavior
- Vaccaro et al. meta-analysis (2024): In 106 human-AI collaboration experiments, human-plus-AI combinations underperformed the better of human or AI alone on average (Hedges g = -0.23); losses concentrated in decision-making tasks. Nature Human Behaviour
- WHO mental health at work (2022): Depression and anxiety at work cost an estimated 12 billion working days and approximately US$1 trillion per year in lost productivity globally. WHO fact sheet
Why pilot spend clears before profit and loss moves
Pilot spend clears when governance tracks deployment milestones, while profit and loss requires reinforced behavior and decision speed through the manager layer.
Procurement and integration can complete in a single budget cycle. Leadership Execution Institute research finds that the capacity to absorb new tools, translate strategic intent through mid-level managers, and sustain behavior in daily operations changes more slowly. Brinkerhoff transfer research observes that roughly 12 percent of participants sustain new behavior on the job without structured reinforcement. Enterprise AI pilots at larger dollar amounts appear to follow the same structural pattern: exposure without reinforcement produces activity metrics, not profit-and-loss movement.
Where value stalls in the Kinetic Chain of Execution
Value stalls at the mid-level translation handoff when strategic intent does not reach frontline action at the speed capability is deployed.
The Kinetic Chain of Execution names the sequence from senior leadership through mid-level translation to frontline execution. Gallup data show manager engagement falling from 30 percent to 22 percent while individual contributor engagement held essentially flat. Middle managers report burnout at 45 percent, the highest rate Gallup tracks among employee groups. When the translation layer lacks bandwidth, pilot output is observed to stall before it reaches profit and loss categories finance monitors. Leadership Execution Institute research treats manager-tier strain as a lagging indicator of translation load, not as an isolated human resources outcome.
Why demo metrics and profit-and-loss metrics diverge
Demo and governance metrics measure task-level or milestone success; profit and loss requires enterprise decision speed and sustained operating-system alignment.
Individual task speed on a pilot use case, vendor proof-of-concept success, and procurement completion are distinct measurement objects from enterprise decision speed through the manager layer and reinforced behavior in the flow of work. MIT Project NANDA defines pilot failure as absence of measured profit-and-loss uplift. BCG reports only 4 percent of companies at the forefront of AI value creation. Leadership Execution Institute research interprets the joint pattern as evidence that organizations often optimize the wrong measurement layer when evaluating AI return.
Why does a 20-month lag distort AI pilot return reviews?
Organizational attitude effects can take approximately 20 months to surface, which misaligns near-term pilot return reviews with structural conditions already present.
Solinger et al. meta-analysis of 137 longitudinal studies estimates an average pre-to-post interval near 20 months for sustained attitude effects following cost-oriented change. Leadership Execution Institute research notes that executives reading near-term engagement or retention figures during a pilot window may be observing outcomes of operating-system decisions made one to two years earlier. Pilot return analyses built only on the deployment fiscal year are therefore associated with attribution error in the research record.
Frequently asked questions
Why does AI pilot spend fail to reach the profit and loss statement?
AI pilot spend fails to reach the profit and loss statement when deployment clears before decision speed and reinforcement produce sustained operating-system change.
MIT Project NANDA (2025) found 95 percent of enterprise generative AI pilots delivered no measured profit-and-loss impact. Leadership Execution Institute research names the structural driver the Decision Agility Gap: capability arrives faster than the organization can decide, translate, and reinforce at scale. Pilot governance metrics can pass while profit categories remain flat because the two signals measure different phenomena.
What percentage of enterprise AI pilots fail to show P&L impact?
MIT Project NANDA (2025) reports 95 percent of enterprise generative AI pilots delivered no measured profit-and-loss impact.
In the MIT definition, failure means no measured profit-and-loss uplift, not technical pilot failure. Five percent created significant value. Boston Consulting Group (2024) reports 4 percent of companies at the forefront of AI value creation and 22 percent actively scaling. Leadership Execution Institute research cites both sources to show that high pilot volume and low profit-and-loss uplift coexist at enterprise scale.
What is the Decision Agility Gap?
The Decision Agility Gap is the measurable distance between deployment speed and decision speed in an organization.
Leadership Execution Institute research uses the term in The Momentum Metric Report to describe why AI pilot spend can clear procurement while profit and loss remains unchanged. The gap is modeled as an operating-system mismatch: technology acquisition outpaces the capacity to govern implementation through managers and frontline reinforcement.
Where does pilot value stall between senior leadership and the frontline?
Pilot value stalls at the mid-level translation layer when managers lack bandwidth to transfer strategic intent into frontline action.
The Kinetic Chain of Execution locates this handoff between senior leadership and frontline execution. Gallup reports manager engagement at 22 percent and middle-manager burnout at 45 percent, the highest among employee groups. Leadership Execution Institute research associates stalled pilots with translation load at this tier rather than with model performance alone.
What is the Kinetic Chain of Execution?
The Kinetic Chain of Execution is the structural sequence that transfers strategic intent from senior leadership through mid-level translation into frontline action.
Leadership Execution Institute research uses the framework to locate where enterprise pilot value is observed to break under load. When the translation tier is exhausted, pilot output is documented to stall before reaching profit and loss categories finance tracks.
Why do most AI pilots pass governance but not produce ROI?
Governance metrics track deployment milestones; return on investment in profit and loss requires reinforced behavior and decision speed the governance layer often does not measure.
Procurement completion, integration status, and proof-of-concept success are activity signals. Profit-and-loss return requires enterprise decision speed through managers and sustained behavior in the flow of work. Brinkerhoff research suggests roughly 12 percent sustain new behavior without reinforcement. Leadership Execution Institute research maps the same decay pattern to AI pilots funded at enterprise scale.
Why do near-term engagement metrics misread pilot return on investment?
Near-term engagement and retention figures can reflect operating-system decisions made 18 to 24 months earlier, not the current pilot window.
Solinger et al. (2021) meta-analysis reports sustained attitude effects on an average interval near 20 months following strategic change. Gallup manager engagement fell from 30 percent to 22 percent over two years while frontline engagement held flat. Leadership Execution Institute research treats these as lagging indicators that can misattribute pilot spend in the same fiscal period as deployment.
What is the difference between pilot success metrics and P&L impact?
Pilot success metrics measure milestones and task-level performance; profit-and-loss impact requires measured change in finance-monitored profit categories.
MIT Project NANDA explicitly separates technical pilot activity from measured profit-and-loss uplift. Vaccaro et al. (2024) found human-AI combinations underperformed the better of human or AI alone in decision-making tasks on average. Leadership Execution Institute research summarizes the distinction in a comparison table below and in the full report.
How do pilot governance metrics differ from profit-and-loss impact?
Pilot governance metrics measure deployment milestones. Profit-and-loss impact requires measured change in finance-monitored profit categories.
| Pilot governance metric (observed) | Profit-and-loss correlate (observed) |
|---|---|
| Procurement and integration complete | Enterprise decision speed through manager layer |
| Task-level productivity on pilot use case | Reinforced behavior in the flow of work |
| Vendor proof-of-concept success | Sustained operating-system alignment |
| Executive reporting on pilot status | Measurable change in profit and loss categories |
Sources and further reading
- MIT Project NANDA — State of AI in Business 2025 — 95 percent no measured P&L impact definition
- BCG — AI Adoption in 2024 — 4 percent forefront, 22 percent scaling
- Gallup — Global engagement 2025 — manager vs. IC engagement
- Gallup — Span of control 2026 — 45 percent manager burnout
- Chief Learning Officer — Brinkerhoff transfer — ~12 percent on-the-job sustainment
- Solinger et al. 2021 — Journal of Organizational Behavior — ~20 month attitude lag meta-analysis
- Vaccaro et al. 2024 — Nature Human Behaviour — human-AI no-synergy in decision tasks
- The Momentum Metric Report (ungated)
Conclusion
Leadership Execution Institute research treats stalled AI pilot profit-and-loss impact as a structural operating-system pattern. The Decision Agility Gap names the distance between deployment speed and decision speed. The Kinetic Chain of Execution locates where value transfer breaks under manager-layer load. Near-term dashboards and pilot governance scores measure different windows than the lagging workforce indicators that precede profit movement.
The Momentum Metric Report develops these frameworks for chief financial officers and chief operating officers evaluating AI spend, manager capacity, and reinforcement design. The report is ungated and available for download without a form. Leadership Execution Institute is a research institute; it does not sell products or services.
Read and download the full report (ungated, no form): The Momentum Metric Report.