# Teams: Comparing Manager Expectations With Employees’ Experience

Compare managers’ expectations with employees’ experience through observable work situations, without diagnosing people or inventing a cost.

- Canonical URL: https://www.harmate.com/en/blog/the-hidden-cost-of-misaligned-teams-and-how-semantic-audit-reveals-it
- Author: Harmate Team
- Published: 2025-12-10
- Updated: 2026-10-18T00:43:48.326428+00:00
- Language: en

## Content

# Teams: Comparing Manager Expectations With Employees’ Experience

Differences between a manager and a team can complicate a task, decision, or handoff. Collected responses can help locate those differences, but they do not calculate a “hidden cost” on their own or measure how compatible people are. To understand a situation, ask both sides to describe the same work situations, then check what happens in practice.

For the specific context of psychosocial-risk (RPS) prevention, the [INRS recommends complementing questionnaires with other sources, such as interviews, work observations, and document review](https://www.inrs.fr/risques/psychosociaux/questionnaires). We borrow only this general principle of combining sources: the protocol and grid below are editorial aids for clarifying work situations, not an RPS-prevention method, validated questionnaire, or diagnostic tool.

If your starting point is instead an aggregated QWL survey score, the [QWL barometer guide](/en/blog/qwl-barometer-why-your-indicators-are-green-while-morale-is-dropping) shows how to keep measured findings, collected accounts, and hypotheses to check distinct.

## Start with a work situation, not a label

Avoid asking “Are you aligned?” or classifying a group as “resistant,” “insecure,” or “a poor fit.” Choose a specific situation instead: setting a priority, reviewing a deliverable, responding to an urgent request, or handing over information. Ask the manager and the people involved separately what they understood, did, and expected.

A mirror questionnaire is useful when it asks comparable questions and leaves room for examples. It is not a way to rate people against each other or infer a psychological trait from a few words.

## A protocol for hearing both perspectives

**1. Set the topic and ground rules.** Define a work-related question, the group concerned, how answers will be used, and who can access them. Explain how findings will be shared. Do not promise anonymity until you have checked identification risks, especially in a small team.

**2. Ask parallel questions.** Ask employees and the manager the same open question separately: “Tell us about the last time an urgent request interrupted planned work. What did you do? What helped or was missing?” Add a question suited to the topic instead of using a keyword list to classify responses.

**3. Compare situations, not people.** Note the event described, the rule each person understood, room to make a decision, and expected result. Keep differing accounts and cases that do not support the first hypothesis. If the topic is “responsiveness,” ask what that means in practice: what response time, which displaced priority, and who makes the trade-off?

**4. Check the difference.** Present findings as points to clarify: “The manager thought urgent work could be prioritized day to day; several responses describe priorities changing without an explicit decision.” Ask the people concerned whether the summary reflects their experience. Compare it with context such as instructions, deadlines, and review arrangements.

**5. Agree on a revisable trial.** Choose one limited adjustment, such as an escalation rule for urgent requests or a review criterion. Decide who will try it and when you will revisit the situation. The review concerns the task and working arrangements, not a score assigned to individuals.

## A grid for reporting observable differences

| Situation | Manager’s perspective | Team’s perspective | Difference to verify | Clarifying question | Possible trial |
|---|---|---|---|---|---|
| An urgent request arrives during planned work. | “The person can decide what to postpone.” | “The planned task is still due the same day.” | Room to reprioritize is unclear. | Who can move a deadline, and how should recipients be told? | Try a written prioritization rule for a defined period. |
| A deliverable needs to be sent. | “A first draft is enough to get feedback.” | “I thought it needed to be final.” | Expected level of completion differs. | What makes a document ready for review? | Add a short review checklist. |

Fictional example: if a manager talks about “taking initiative” and team members ask for more approval, do not conclude that the team lacks autonomy. Check which decisions are actually delegated, what risks apply, and what feedback followed earlier initiatives.

Do not reduce several accounts to a dramatic quote. Keep context, state what remains uncertain, and do not present an internal frequency as representative of the organization. If an answer raises a health, violence, or occupational-risk issue, use the appropriate response process rather than treating it as a simple relationship disagreement.

## Move from observation to action

After the discussion, report the situations reviewed, descriptive areas of agreement and difference, what still needs checking, the decision made, and its review point. If the team measures time spent, rework, or delays, first define the data, period, and scope; do not apply a generic percentage to a financial cost. A change observed after a trial does not by itself prove that the trial caused it.

For a further method for reading a set of written responses, see [reading 200 open responses without fooling yourself: a method for deciding without drowning the signal](/en/blog/analyzing-200-open-responses-without-bias-a-method-for-actionable-decisions). In Harmate, [Open Questions](/en/product/open-questions) supports reading written responses with their source excerpts so the team can interpret them. The team decides whether any excerpt is relevant to its debrief.

An observed difference can become a useful question about work. It is not a personality diagnosis, compatibility measure, or promise of financial performance.