And Another Thing... Why Engagement Surveys Miss Retention Problems

Angelia Williams Graves, Founder of Plinth Advisors and Virginia State Senator

Engagement surveys measure how employees feel about an organization. They do not

measure whether people know who is responsible for what. A team can report high satisfaction while losing hours each week to work that gets done twice. Gallup estimates replacing one employee costs 50-200% of annual salary, a cost engagement scores rarely predict.

The survey comes back, and the numbers are fine. Participation is up. Satisfaction holds steady. Manager scores are respectable. And somewhere in the presentation, a senior leader asks the question everyone is thinking: if the numbers are this good, why doesn't it feel this good?

Most chief human resources officers have lived that moment. They know there's something the instrument can't capture, and the language for it is hard to come by, which makes funding it very difficult.

What do engagement surveys actually measure?

Engagement surveys measure sentiment: satisfaction, belonging, perceived fairness, confidence in leadership. These are real and worth knowing.

What they do not measure is behavior. Whether people define ownership the same way. Whether a slipping commitment gets named before the deadline or after it's gone. Whether feedback is received as information or as a verdict.

An employee can genuinely like their colleagues, believe in the mission, feel fairly compensated, and still spend hours each week redoing work that came back unfinished because two people each thought the other one had it. That employee will score you well. But engagement surveys never ask the question that would catch the real problem.

What four questions surface the gap a survey misses?

These four questions get at behavior rather than sentiment, and they are not on standard survey instruments:

1. Does everyone here define ownership the same way?

2. When a commitment slips, does anyone say so before the deadline rather than after?

3. Can a person on this team receive difficult feedback as information rather than as a verdict?

4. Would two people on this team describe the standard for good work in the same words?

Sentiment does not determine output. Behavior does.

Why are exit interviews unreliable data?

Exit interviews are among the least reliable data an organization collects because departing employees have almost no incentive to be honest. They need a reference. They may want to leave the door open. So they say it's an opportunity for advancement, which is true enough to say out loud and vague enough to end the conversation.

The stated reason gets logged, the role is backfilled, the underlying conditions remain, and the replacement leaves within a year for the same unrecorded reason.

The Bureau of Labor Statistics counts roughly 3.1 million voluntary quits per month in the United States. Organizations read that as a labor market condition, something happening to them. A meaningful portion is a behavioral condition, something happening inside them. The distinction matters, because the first cannot be changed and the second can.

What does the professional character gap cost an organization?

Gallup estimates the cost of replacing a departing employee at 50-200% of that person's annual salary. For a single mid-level professional, that is a six-figure loss before recruiting posts the job.

That is the visible cost. The invisible cost accumulates in the months before the departure:

• Meetings held twice because the first decision did not hold

• Managers who stop delegating after three incomplete handoffs

• Strong performers absorbing unassigned work until absorbing it stops being worth the paycheck

• Deadlines that slide by two or three days so routinely that the slide becomes the schedule

None of that appears as a line item. All of it appears in output.

The gap is also wider than most organizations believe. NACE research finds a difference of close to 30 percentage points between how employers rate the professionalism of their newest hires and how those hires rate themselves. Two groups evaluating the same person and reaching different conclusions. Neither is being dishonest. They are working from different definitions of competent, and nobody has put the definitions side by side.

Why doesn't more training close the gap?

Most leadership and professional development spending assumes the behavioral floor is already in place and builds upward from it. Strategic thinking, executive presence, influence without authority. The content is good. The foundation was never poured.

There is also a targeting problem. Development aimed only at emerging professionals carries an implicit message: this standard applies to you, and the people above you have already cleared it. That is rarely true, and everyone in the room knows it. It is why this work so often produces compliance instead of change.

Leveling the professional character playing field means one standard applied in both directions at once. Team members and leaders are held to the same definition. That is what makes it credible, and credibility is what makes it hold.

What can a CHRO do before the next survey cycle?

No new instrument is required. One diagnostic exercise will surface more than another survey round.

1. Identify your last three missed deadlines. The ordinary ones, not the catastrophic ones.

2. For each, locate the moment where someone assumed a task belonged to someone else. It will be there almost every time.

3. Ask whether anyone on that team could have named the owner, in writing, before it slipped.

If the answer is no, the gap is not motivation. Your people are trying.

The Professional Character Cost resource walks through what that gap is costing your organization and includes five conversation suggestions you can use this week.

Gartner's 2026 survey of chief human resources officers named leader and manager effectiveness the top priority of the year. The organizations that move the needle will not be the ones that buy the most. They will be the ones willing to define the floor everyone stands on and hold everyone to it, beginning at the top.

Your engagement survey will tell you how your people feel. That is useful. It is not what is costing you money.

Look at your last three misses this week. See whether ownership was ever actually named. Then decide what your standard will be.

Frequently Asked Questions

Do engagement surveys predict employee turnover?

Engagement surveys predict turnover weakly. They measure how people feel about the organization, not whether the team knows who is responsible for what. A satisfied employee can still spend hours each week redoing work because two people each thought the other one had it. Standard surveys don't contain questions that would catch it.

What is the professional character gap?

The professional character gap is the distance between the behavior an organization assumes its people share and the behavior they actually bring to work. It shows up when nobody is sure who is responsible for what, when a commitment made on Monday quietly dies by Thursday, and when feedback comes across as an insult instead of information. It is structural, not personal, and it can be taught.

How much does employee turnover cost?

Gallup estimates that replacing a departing employee costs 50-200% of that person's annual salary. For a mid-level professional, that is a six-figure loss. The figure does not include the quieter costs: work that has to be done twice, the same meeting held twice, and tasks strong performers absorb because no one else claimed them.

Why do exit interviews fail to identify real reasons for leaving?

Departing employees have little incentive to be candid. They may need a reference or want to preserve the relationship, so they cite "opportunity for advancement" rather than the real reason. The organization logs an inaccurate reason, backfills the role, and repeats the loss with the replacement.

What should replace or supplement an engagement survey?

Supplement rather than replace. Look at your last three missed deadlines and find the moment where someone assumed a task belonged to someone else. Then ask whether anyone on that team could have named the owner, in writing, before it slipped. That one exercise shows you gaps a survey cannot, and it costs nothing.

The base that elevates everything.