A CEO I know sat through a quarterly review last spring. The numbers looked fine on paper. Revenue up. Headcount up. Customer churn down. Then a director raised her hand and asked why their best engineering team had quietly lost three of its five senior people in nine months, all to the same competitor. Nobody in the room had seen it coming. The exit interviews said all the right things. New opportunity, career growth, closer to family. The data told a different story, and nobody was reading it.
That story is not exceptional. Gallup has tracked for years that only about a third of US workers feel engaged at work, a figure that has barely moved despite record spending on people programs. Meanwhile, the cost of replacing a highly skilled employee can run anywhere from half to two times their annual salary, according to the Center for American Progress. The money is real. The question is why leaders keep flying blind until the resignations land in their inbox.
The answer is that most companies measure the wrong thing, on the wrong cadence, with the wrong tool. Annual or quarterly surveys produce snapshots that are stale by the time the report renders. We wrote about this exact cadence problem and what it costs your decisions. And the frame itself is off. Leaders do not need another opinion poll about whether people are happy. They need to see how work actually happens, in real time, across the organization.
That is what we call organizational telemetry.
Start With the Work, Not the Feelings
Organizational telemetry is the continuous measurement of how work flows through a company. It tracks collaboration patterns, decision velocity, handoff quality, and the friction that builds up in the gaps between teams. It is the difference between asking people how they feel and watching what they do.
The distinction matters because behavior does not lie the way self-report does. A mid-quarter conversation with your head of engineering tells you what happened last week through the lens of someone who wants to sound competent. Telemetry tells you that a critical dependency has crossed twelve people and two time zones in four days without a decision being made.
The behavioral evidence is strong. Project Aristotle at Google found that psychological safety, not individual talent, was the strongest predictor of high-performing teams. That finding is years old and it has not aged a day. But here is the trap. You cannot ask people to reliably report their own psychological safety and expect an unbiased number. They answer through the same lens of social desirability that clouds every self-assessment.
Cultural Intelligence Turns Noise Into a Decision
Telemetry gives you the raw signal. Cultural intelligence is what turns that signal into something you can act on. It is the read on how your organization actually behaves under pressure, how norms form, who gets heard, and where the real bottlenecks to working well live. We have written about how to put cultural intelligence to work and why it matters for business results.
This is where the old tools fail hardest. A static survey asks a narrow question once and calls it done. Cultural intelligence is continuous and contextual. It connects a drop in cross-team collaboration to a restructuring that happened two months earlier, or a spike in rework to a process change that nobody signed off on, and surfaces the causal line. That is what moves you from opinion to insight.
The evidence that this pays is well documented. McKinsey has found that companies in the top quartile for healthy organizational practices outperform their peers by a wide margin on total shareholder return. Deloitte has framed culture as a competitive advantage for years. When a company behaves differently, it compounds, good or bad.
The Cost of Flying Blind Is Not Theoretical
Consider the actual math of a bad decision made in the dark. A single avoidable resignation of a senior engineer costs six figures in recruiting, ramp time, and lost institutional knowledge. A failed change initiative, which Gartner puts at a failure rate the majority of the time, burns the budget and the credibility of the leadership team that championed it. Both are culture problems wearing business clothes.
Gallup’s long-running business-unit research links low engagement to lower productivity, higher absenteeism, and higher turnover. Steady declines in those indicators are not a soft issue for the people team to handle quietly. They are a leading indicator that org-level averages actively hide.
That lead time is the whole argument. If you only discover the problem after the resignations and the missed quarter, you are doing archaeology, not management. Telemetry gives you the chance to see the decay while there is still time to act on it.
Make It a Discipline, Not an Event
The organizations that get this right treat organizational telemetry as an operating rhythm, not a quarterly event. They watch collaboration health the same way they watch burn rate. They set thresholds. When a metric crosses a line, it triggers a conversation, not a report that gets filed.
The discipline pays off in specificity. Instead of telling your head of people to improve the culture, you can point at the exact three teams where handoff quality has decayed and ask what changed. Instead of a vague concern about retention, you can see which specific roles, which managers, and which moments in the employee lifecycle drive exit risk. People data deserves the same discipline as financial data. That is the difference between management and guesswork.
Start Where the Friction Actually Lives
You do not need to boil the ocean on day one. Pick the two or three metrics that already matter to your operating plan and instrument those. Watch the flow from sales to delivery. Watch the handoff between product and engineering. Watch onboarding velocity, because that is where new people broadcast the real experience of your company before anyone has taught them the talking points.
Run it for a quarter and you will have something the annual survey ritual never gives you, a baseline you can actually move. Then you can ask the version of the question that matters to a board and get an honest answer.
Here is the provocation worth sitting with.
Your last ten hires were made on behavioral data, on references and past performance and observed patterns of work. Why would you manage the two hundred people they joined on anything less?
What would change in your next operating review if you could see how work actually moves through your company, before the resignations and the missed quarter tell you the bad news?
Leave a Reply