I sat in a restructuring meeting where leadership was preparing to eliminate roles, redraw others, and shift compensation under real financial pressure.
The meeting had a cost-savings model. It had action items. It had a plan for staff conversations after the town hall. But the work kept returning to a question the record could not answer: whether the organization had corrected the revenue and cost structure that made the restructuring necessary.
That is the distinction between an accountability record and an accountability system that can tell whether the work changed anything.
A record can show that leadership met, roles changed, messages went out, and a plan was followed. It cannot, on its own, show whether the new structure can carry the work profitably, whether the underlying decisions have changed, or whether the same pressures will bring the organization back to the same room.
The gap is architectural. A measurement system built to confirm the wrong object.
The Record That Was Always Complete
Planview's 2025 Global Benchmark Report documented organizational confidence in adapting to a changing environment dropping from 40% in 2021 to 28% in 2025, alongside faster execution. This is a period of faster execution cycles, more project management infrastructure, and more measurement technology than any prior era. Organizations ran faster. The gap between activity and outcome widened.
Don Sull's MIT Sloan source, drawing on surveys of more than 400 CEOs, identifies execution as the number one strategic leadership challenge and reports that between one-half and two-thirds of organizations fall short of their stated strategic objectives. Bridges Business Consultancy's research found that only 7% of respondents rated their strategy implementation as excellent, while 48% failed to reach even half of their stated strategic targets.
These organizations were executing. Documenting activity. Completing action items. Running the review cadence. And producing a gap between the activity record and the consequence they had committed to.
Harvard Business School's Robert Simons cites Economist Intelligence Unit research estimating that companies lose approximately 37% of their revenue and profit potential through poor strategy execution. Mankins and Steele's 2005 research, cited separately in the paper, found that companies delivered 63% of promised financial performance. A strategy can be architecturally sound while the system that is supposed to hold it confirms only that execution activity is occurring and leaves consequence unmeasured.
What the Institutional Precedent Reveals
The compliance architecture that followed Enron and WorldCom offers an institutional parallel. Sign-off chains at each stage of material decision-making. Internal certification requirements for senior officers. Documentation protocols with specific retention windows. Independent audit trail requirements. Board oversight structures formalized and subjected to independent verification.
The measurement infrastructure confirmed presence at each layer. Document filed. Signature obtained. Process recorded. Certification complete. The regulatory framework verified that governance activity had occurred, with specificity and consequence attached to each step. By every metric the compliance system was designed to produce, the organizations that built it were producing compliance.
Those records established that specified governance activity had occurred. They did not, by themselves, establish whether governance had changed how consequential executive decisions were made inside an organization.
The parallel is not that compliance activity is empty. It is that evidence of compliance activity does not, by itself, prove a change in decision-making. A framework designed to verify specified governance activity can establish that activity with precision while leaving a separate question unanswered: what changed in the operating system because the activity occurred.
This is the measurement-object problem in institutional form. The system confirms the object it was designed to confirm. Activity is confirmable. It produces documentation, timestamps, sign-offs, filed records. Consequence is harder to confirm because consequence exists at the level of what the operating system does differently after the accountability work is complete. When measurement is designed, it is designed around what can be confirmed. What can be confirmed is activity.
The Architecture of the Gap
Meeting churn persists because the accountability system is doing its job. Its job is to confirm presence. It is confirming presence. What the system was not designed to do is confirm what presence produced.
Three patterns characterize where measurement has been pointed at the wrong object in organizations operating at this condition.
The reporting cadence was built around completion. An action item is closed when it is reported complete. Whether the completion produced the change it was intended to produce is outside the closure definition. The system has no mechanism for distinguishing completion that changed something from completion that produced a record of having tried. Both read as complete.
Accountability is structured around attendance at the review. The recurring meeting is the accountability structure. Attendance confirms that accountability occurred. What the attendance produced is not measured as part of the accountability structure. The meeting is the deliverable, not what the meeting changed.
No consequence attaches where measurement is absent. Where measurement is pointed at activity rather than consequence, the gap between activity and consequence is invisible to the system. An organization cannot confirm what it did not build the infrastructure to confirm. The blind spot is outside the system's field of view.
The incentive lock keeps measurement pointed at activity. Completion is binary. Consequence requires a standard of evidence. The organization has to define what consequence looks like before it can confirm that consequence occurred. Most organizations have not built that standard. The measurement defaults to what the system already knows how to confirm. The system already knew how to confirm activity.
What Holds Outcomes Across Time
The organizations that produced outcomes durably across leadership transitions and major restructuring cycles built consequence into the operating system rather than into the person who drove the initiative. When consequence lives in the person who built it (in their relationships, their authority, their institutional knowledge of the organization), it leaves with them. The accountability structure outlasts the person; the consequence it produced does not.
Building measurement architecture that holds consequences requires re-pointing measurement from what is easy to confirm to what needs to be confirmed. That is a design choice. It requires deciding, before the accountability structure is built, what evidence would confirm that consequence has occurred.
For a restructuring commitment, that means confirming the new governance structure was documented and signed off on, then confirming that decision-making in the domain it governs now originates differently than it did before the restructuring. For an execution commitment, it means testing how completed action items changed the relevant operating system.
The re-pointing requires the organization to define what consequence looks like in advance, before the accountability cadence begins. The activity-confirmation system never required this. An accountability system that was designed before consequence was defined will confirm whatever it was designed to confirm. If it was designed around completion, it will confirm completion. The consequence question arrives after the system is already running. At that point, the system reads it as an anomaly rather than an input.
The question is whether the organization can show what changed because the accountability process ran. That is the point at which a complete record stops being enough evidence.
A related whole-organization read is Independence assumed before the function exists.
ORBIT — Organizational Readiness, Bottlenecks, Infrastructure, and Traction gives you a first read of the operating system underneath the strain.
If your accountability system disappeared tomorrow, what evidence would remain that the outcomes it tracked had actually changed the organization?