Your Retention :Plan Runs Through the People Most Likely to Leave

Two pieces of research landed this month, and separately they read like ordinary bad news. Put them next to each other and they describe something worse.

Eagle Hill Consulting's Employee Retention Index fell in the second quarter to its lowest point in twelve months. The pressure is coming from compensation. People are frustrated by what they are paid and optimistic about what they could get somewhere else, and with 7.6 million openings sitting in the JOLTS data, that optimism has somewhere to go.

Here is the part I have not seen anyone pick up.

The steepest decline belongs to Millennials. They also reported the lowest satisfaction on organizational confidence, on pay, and on culture, and the highest confidence that they could find something better. Eagle Hill's own report puts the consequence plainly: "As Millennials increasingly occupy management, leadership, and specialized professional roles, this shift could have outsized implications for organizations."

Read that again with your org chart in front of you. The group showing the largest retention decline is the same group you are counting on to hold everyone else. Your retention strategy runs through people who are themselves halfway out the door.

The instinct is to fix a compensation problem with compensation. For most companies that door is shut. Budgets were set last fall, the market has 7.6 million openings in it, and a bidding war against that is one you lose slowly and expensively.

Melissa Jezior, Eagle Hill's CEO, warned against reading a slower hiring market as permission to relax. She described workers as sending "a nuanced message." I would put it less gently. People are staying because the market is awkward, and they are telling you they will move the moment it is not.

So the lever is the manager. That is the one that has always worked, it is the one that costs nothing on the comp line, and it is the one we have systematically failed to equip.

Which brings me to the second piece

Sophie Wilson wrote a piece for UC Today in June asking whether engagement fails because nobody owns it. Her argument is structural rather than moral, and she is right.

The numbers she assembles are familiar and they still land. Gallup has 51% of employees disengaged. Gallup also finds that managers account for up to 70% of the variance in their team's engagement scores. Deloitte lists organizational ambiguity among the top drivers of engagement decline.

Her conclusion is that when HR, leadership, and managers all nominally own engagement, nobody is structurally accountable for it, and shared ownership without defined deliverables collapses into no ownership at all.

I spent twenty years in HR and several more in the CEO seat, and I have watched that exact collapse from inside both. Engagement lives in the HR deck. Accountability lives in the manager's team. The two never meet in a performance review.

Fix the ownership and you still have a problem, which is the one I keep running into.

Say you do it properly. You write engagement into your directors' goals. You tell them, in writing, that the health of their team is their job. Good. Now hand them the instrument you have given them to do it with.

It is a survey. It went out in July. The results came back in October. They describe how a team felt about a manager's behavior in a set of conversations that happened three months ago, aggregated into a score, stripped of anything that would let you locate the moment things went wrong.

We have assigned accountability at the speed of a conversation and delivered the data at the speed of a quarter.

No other function tolerates that gap. A sales leader carrying a number gets the pipeline daily. An engineering lead who owns uptime gets paged in the moment. In people work we hand over the outcome, deliver the evidence a quarter later, and call it a program.

There is one Gallup finding in Wilson's piece that should end the debate about whether this is worth fixing. Employees who see no action taken after an engagement survey are 74% more likely to be actively disengaged within twelve months.

That number reframes the whole exercise. The survey is not a neutral instrument that either helps or does nothing. Run one and let it die in a slide deck, and you have manufactured disengagement in people who were fine before you asked.

Combine that with the Eagle Hill data and you get the situation most companies are walking into this fall. Retention is softening. The people it is softening under are your managers. You have asked everyone how they feel, produced nothing they can see, and made the population you depend on measurably more likely to leave.

Three things, in order of how hard they are.

1. Name the owner and write it down. Wilson is right that ambiguity is the root. If engagement belongs to your managers, it goes in their performance review with a number attached, the same way any other accountability does.

2. Give them the capability. Accountability handed to someone who was promoted for being good at the job rather than good at the people is a trap you set yourself. Most managers have never been taught how to run a hard conversation. Telling them they own the outcome does not teach them.

3. Close the latency. This is the one almost nobody is working on, and it is the one that decides whether the first two matter. A manager who owns the health of a team needs to know how a conversation landed while there is still something to do about it. Not next quarter. That afternoon.

The manager is the highest-leverage instrument in the building, worth 70% of the variance, and the only feedback we have ever given them arrives after the person they lost has already given notice.

That last gap is what pulled me back in. After twenty years of building people programs and one very long lesson in what happens when the data about your people sits in a box, I wanted to work on the part where somebody finds out in time to do something.

That is what we are building at Relate. If you want to see it, come find me.