The Executive Operating Model · Part 4 of 6
Measuring Execution Health
Commitment accuracy alone is easy to misread. Execution health takes a small set of honest signals, read together on the sprint cadence, to show whether follow-through is holding or eroding.
The executive sprint produces something the earlier horizons could not: a set of specific, owned outcomes committed for the month. The previous piece ended on the first thing it produces, which is a pair of numbers. Commitment accuracy asks how many of the outcomes a team committed to it actually finished, and roll-forward asks how much of what it committed slid into the next month. Those two ratios are the beginning of an answer to whether the model is working, and on their own they are also easy to misread. A team can post a high commitment-accuracy number by committing to very little, or by quietly narrowing an outcome late in the sprint so that whatever landed still counts as done. Read in isolation, a single ratio tells you almost as little as the year-end impression it was meant to replace.
What the executive team needs is a small instrument set, read together on the sprint cadence, that shows whether follow-through is holding or eroding while there is still a quarter left to act on it. Execution health is measurable in the same sense delivery health is, through a handful of honest signals the team commits to reading and acting on rather than a single scoreboard number. The purpose of measuring is to convert a vague sense that the year was busy and somehow behind into evidence specific enough to change what the team does at its next rebase. This piece describes that instrument set, the discipline that keeps it honest, and the reason the signals matter far more as a pattern than as a report.
The signals the model reads
The model reads execution health through two families of signal, because two different things can go wrong. The first family reads whether committed work is actually being completed. Commitment accuracy is the share of committed outcomes a team finished. Roll-forward rate is the share of committed work that moved to the next sprint. Median initiative age is how many sprints an active initiative has stayed open, which catches the high-value work that never quite closes. Time to strategic outcome is the elapsed time from starting a pillar-aligned item to finishing it, watched for whether it trends down. Percent of annual plan advanced is the share of the year’s priorities that moved this sprint. Together these read the output of the cadence: whether outcomes are finishing, whether they are aging, and whether the annual plan is genuinely advancing month by month rather than in a year-end rush.
The second family reads the conditions that produce that output. Exec pulse is a simple self-rating from one to five of whether a leader felt focused and impactful during the sprint. Priority churn is the share of a sprint’s work that was added mid-month and was never in the plan. Dependency resolution is the share of known blockers cleared within the sprint. Strategic drift time is the share of executive time consumed by unplanned or misaligned work, drawn from the drift log described below. Where the first family tells you whether the plan advanced, the second tells you whether the executive team had the focus and the clear path required to advance it. A team can hold acceptable completion numbers for a month or two while its focus signals are already deteriorating, and the focus signals are usually the earlier warning.
Each metric carries a target band, so the number prompts an action rather than merely describing a state. Green is healthy, watch warrants a conversation at the sprint review, and investigate signals that the model itself needs correction. Commitment accuracy at or above eighty percent reads green, seventy to seventy-nine is watch, and below seventy is investigate; strategic drift time at or below fifteen percent is green and above twenty percent is investigate. The discipline that matters most here is what the bands are and are not. They are proposed starting points, offered to be calibrated against a team’s own baselines over the first few quarters, and they are not benchmarks drawn from a study or thresholds that certify a team against anyone else. A band’s only job is to turn a raw figure into a prompt to talk or to act, since a number without a band is a figure no one has decided how to use, and every team should expect to move the boundaries as it learns what healthy looks like for its own work.
Of all these signals, strategic drift is the one most likely to be misread, and how it is captured decides whether it helps. The Strategic Drift Log records the work that falls outside planned commitments: the fire drill, the unplanned initiative, the request that arrived with a name attached and could not easily be refused. Each entry notes who did the work, roughly how much time it took, why it happened, and what it displaced. It is self-reported and reviewed once a month at the sprint review, alongside the numbers.
The instinct is to treat a drift log as a productivity tracker, a way to catch executives spending time on the wrong things. Used that way it would be filled out defensively and would tell you nothing. Its purpose runs the other direction. Logged honestly, drift exposes structural causes rather than personal failings. A recurring fire drill from the same account points at an ownership gap. A steady stream of unplanned hiring work points at a staffing shortfall. A pattern of reactive requests points at demand the operating model has not accounted for. The individual entry matters less than the pattern across entries, which is why the review reads the log for repeated causes and repeat sources rather than tallying hours against people. Treated as a signal about the system, the drift log protects executive focus; treated as a scorecard on individuals, it destroys the honesty that makes it worth keeping. Holding it in the first category is a choice the team has to make deliberately every time it reviews the log.
Reading the signals as a set
No single one of these signals should trigger a decision on its own, because each has an innocent explanation. Commitment accuracy can dip in a month that carried unusually ambitious outcomes. Drift can spike for one genuine emergency. Pulse can fall because two leaders were traveling. The measurement loop earns its place when the signals are read together, because a real execution problem shows up as several of them moving in the same wrong direction at once.
Consider a hypothetical quarter, offered only to show how the reading works and not as a reported result:
| Signal | Month 1 | Month 2 | Month 3 |
|---|---|---|---|
| Commitment Accuracy | 90% | 83% | 65% |
| Strategic Drift Time | 15% | 23% | 28% |
| Exec Pulse (1 to 5) | 4.2 | 3.9 | 2.9 |
| Priority Churn | 8% | 12% | 18% |
| Dependency Resolution | 95% | 88% | 91% |
Across the quarter, three signals move together in the wrong direction. Commitment accuracy falls from ninety percent into the investigate band at sixty-five, exec pulse drops from a healthy 4.2 to 2.9, and strategic drift time nearly doubles from fifteen to twenty-eight percent, with priority churn climbing alongside them from eight to eighteen. Read as a set, the story is legible: unplanned work is crowding out committed work, the executives feel it, and follow-through is eroding as a result. The one signal that holds steady sharpens the reading further. Dependency resolution stays between eighty-eight and ninety-five percent throughout, which says the team can still clear blockers when it focuses. What it has lost is the focus itself, as capacity is pulled away before commitments can land. That is the difference between a scoreboard and an instrument: read as a set, these signals point to the cause rather than only registering that something slipped.
The value of reading the set this way is timing. This pattern is visible in the third month, a quarter into the year, while there is still room to respond, rather than surfacing as a December sense that the team was overloaded and behind. That is exactly what the quarterly rebase exists to act on. Entering the next quarter, the team can interrogate the drift log for the recurring causes, protect capacity by deferring or delegating lower-priority commitments, and re-sequence the coming sprints to restore accuracy before drift becomes the norm. The measurement loop is what makes the rebase a response to evidence, and the rebase is what makes the measurement worth collecting; without the signals the team re-sequences on instinct, and without the rebase the signals become only a more precise way to feel bad after the fact.
Reading well also means resisting the single status color. An initiative that is on track to ship can still be the wrong initiative, so the model reads each one on more than one axis: delivery health, whether it is on track, at risk, or off track; strategic health, whether it is still high value or worth re-evaluating; and the leadership attention it needs, whether to monitor it, intervene, or escalate. A green delivery light on an initiative that is no longer worth doing is precisely the thing a single indicator hides. In the same spirit, because executive overload is usually invisible until someone reaches the end of their capacity, the model tracks how executive time is spread across categories such as strategic execution, operational leadership, firefighting, and people leadership, and watches those proportions shift over quarters. Rising firefighting against falling strategic-execution time is the same story the drift log tells, seen from another angle, and the two together make a case no year-end reflection could assemble.
Assembled, these signals give the executive team a reading of its own execution it has rarely had before: a small set of measures the team reads on itself, on the same cadence it commits and delivers work, rather than a report it sends upward. What working looks like becomes concrete and checkable. Over successive quarters, committed outcomes are completed more often than not, roll-forward shrinks, quarterly priorities stop surprising the organization, executive meetings get shorter and more decision-focused because the status is already visible, cross-functional blockers clear faster, and strategic drift trends down rather than up. When those indicators hold, the character of the strategy discussion changes. It shifts from explaining why last year’s execution fell short toward deciding where to aim next.
None of this measurement makes a team execute. It makes execution legible, early enough to steer, which is the entire and deliberately modest ambition of the loop. The signals do not run the company, and a team that turns them into an elaborate reporting apparatus will have rebuilt the overhead the model was meant to remove. The measurement has to stay as light as the cadence it observes, which is a discipline of its own and the last thing the model has to get right: how to install this operating model and keep it lightweight as the organization grows, instead of letting it harden into the bureaucracy it was built to replace. That is where this series goes next.