When an organization invests in coaching for managers, asking whether it makes a difference is reasonable. Asking what managers discussed with their coaches is a different matter.

The distinction should shape the evaluation from the beginning. Employers need useful evidence about progress. Managers need room to examine uncertainty, question their habits and discuss difficult relationships without turning every conversation into a report for their employer.

A practical approach is to evaluate what changes at work while keeping the conversations behind those changes confidential. This requires clear objectives, proportionate measurement and explicit agreements about what will be shared.

Start With Observable Changes

“Become a better leader” is an aspiration, not a useful evaluation criterion. Before coaching begins, translate broad ambitions into two or three behaviors that colleagues can reasonably observe.

For a manager working on delegation, this might mean agreeing to responsibilities before work starts, clarifying which decisions team members can make independently and reducing unnecessary approval steps. For someone developing their feedback skills, it might mean addressing issues promptly, using specific examples and agreeing to a next step.

The manager, their line manager and the coach can discuss these objectives together, while preserving the manager’s ownership of the coaching process. Shared workplace objectives do not require every personal concern explored during coaching to become a shared objective.

A useful starting question is: “If this program helps, what would people notice us doing differently?”

Define the Boundaries Before Collecting Data

Confidentiality should be designed into the evaluation rather than added as a reassuring sentence at the end.

Before the program starts, participants should understand what information will be collected, who will receive it, why it is needed and how long it will be retained. The agreement should also explain the limits of confidentiality and how any necessary disclosure would be handled.

Keep three categories separate: program administration, progress towards agreed upon objectives and session content.

Administrative information might include whether sessions took place. Progress information might cover changes in agreed workplace behaviors. Session content includes the concerns, relationships and reflections explored with the coach.

Participation in an employer-funded program should not be treated as blanket permission to share this third category. Any proposed disclosure needs to be considered against the coaching agreement and applicable obligations.

Build a Small, Repeatable Measurement Process

An elaborate dashboard is not essential. A few relevant questions, asked consistently, can provide a clearer picture than a long survey disconnected from the program’s objectives.

Begin with a baseline. Ask the manager to assess the selected behaviors and describe a recent example. Where appropriate, invite their line manager or team to answer corresponding questions about what they observe.

For delegation, a useful statement might be: “I understand which decisions I can make without seeking further approval.” This is more specific than asking whether someone is a good manager.

Repeat the questions at an agreed point near the end of the program, with a later follow-up to explore whether changes have lasted. Choose timing that gives managers a realistic opportunity to practice.

Keep the process proportionate. Collecting more information is only worthwhile if it helps answer a relevant question.

Compare Perspectives Without Turning Feedback Into a Verdict

Self-assessment, feedback from colleagues and operational indicators offer different perspectives. None provide a complete account on its own.

A manager may feel more confident while their team has yet to notice a difference. A team may report clearer priorities even though delivery remains affected by staffing shortages. These differences deserve investigation rather than being averaged away.

Ask what changed, what remained difficult and what else happened during the program. A reorganization, a new colleague or a change in workload may influence the same outcomes.

An improvement following coaching does not, by itself, demonstrate that coaching caused it. Reports should distinguish observed changes from claims about attribution. Where the evaluation cannot isolate coaching’s contribution, say so.

Report at the Right Level

Different audiences need different information. A manager may benefit from detailed feedback on their own objectives. A program sponsor may need an overview of participation, progress and improvements to the program’s design. Neither purpose automatically requires access to coaching notes.

For program-level reporting, use aggregated findings where appropriate and consider whether individuals could still be identified. Removing names is insufficient when a quotation, job title or distinctive situation reveals who is being described.

In small groups, it may be better to omit a breakdown or report less detail. Free-text comments also need careful handling: a memorable example can reveal more than its author intended.

A Practical Example: Delegation

Consider a fictional manager whose team repeatedly waits for approval on routine decisions. The coaching objective is to improve delegation.

Before the program, the manager and team assess whether decision-making responsibilities are clear. During the program, the manager experiments with explicit decision boundaries and agreed check-in points.

At the review, they repeat the assessment and examine examples of decisions handled independently. They also discuss whether less oversight has created confusion or additional rework.

The organization can evaluate these changes without knowing whether coaching conversations explored perfectionism, fear of failure or a difficult past experience. Those discussions may matter to the manager, but they are not necessary evidence for the sponsor.

Make the Evaluation Useful

The purpose of measurement should be to inform a decision: continue the program, adjust its design, improve organizational support or reconsider its objectives.

Session attendance shows participation. Satisfaction shows how participants experienced the program. Alone, neither establishes that workplace behavior improved.

A credible evaluation combines clearly defined objectives, observable evidence and honest limits. The aim is not to reveal what happened inside the coaching conversation. It is to understand whether something useful changed outside it.