The ritual will be familiar. The engagement report lands, the number is down a point, and a meeting gets scheduled about the number. An initiative is born, perhaps a new survey, a new channel, a listening tour. Nothing about how the work happens changes, and next quarter the number is down another point. An entire management routine has grown up around a metric that does not move the thing it claims to measure.

The engagement survey was supposed to make culture manageable at last: quantified, dashboarded, owned by someone. Gallup's own recent data undermines that promise. Engagement has slid to a five-year low across two consecutive years while the broader economy and corporate results have stayed resilient, and even engagement's strongest advocates are now asking in public whether the industry has been measuring the wrong things.

Take that question seriously, because it only has two answers. Either the metric does not capture what matters, or what it captures does not drive the business. Neither answer justifies the current spend.

Consider what an engagement survey actually records: how people feel on a particular day, filtered through whatever is in the air that week. In a year of rolling layoffs, the filter is fear, and fear distorts the instrument in both directions. Some people perform an optimism they do not feel, while others vent a frustration that passes by Friday. Sentiment is real, but it is not the same thing as capability. One team can feel perfectly fine while being slow, poorly coordinated, and steadily losing its best people. Another can feel stretched and be the sharpest unit in the building. Which of those distinctions does your dashboard actually show you?

There is one finding in the engagement literature that reliably holds up, and its nature is worth noticing. Managers account for roughly 70% of the variance in their team's engagement, and manager engagement has fallen hardest of all. Read that finding for what it is: a statement about what a specific person does with a team on a Monday, not a statement about a dashboard. The predictive signal was behavior all along.

The problem was never measurement as such. The problem is measuring the residue instead of the engine. If you want numbers that predict something, watch how the work moves rather than how people feel about it. How long does a real decision take, and does it stay made or come back around for a third round? Do hand-offs land cleanly, or do they generate rework? Did the person who did the best work this quarter get named, or did the credit pool around whoever spoke the most? These things are observable, and unlike a sentiment score, they are fixable.

The next time your engagement number drops, ask one question in the review meeting: which specific behavior changed? If nobody in the room can answer, then the dashboard is giving you the temperature and nothing about the cause, and there are far cheaper ways to learn that morale is down.

o o o o o o o o 

Performance First is a KOhesion series. Kohesion makes team performance visible at the level of behavior, which is the level where it can be changed.