A Better Performance Review Process That Actually Changes Behaviour

Most performance reviews do not change much

The usual performance review has a familiar rhythm.

You book an hour. The employee writes a self-review. The manager writes feedback. Everyone says the right things. A document gets saved somewhere. Then the same patterns return three months later.

That is not always because people do not care. Often, the process is wrong.

A review is treated like a checkpoint, when it should be a growth conversation. The goal is not to complete the form. The goal is to help both sides see reality more clearly and agree what needs to change next.

Research on feedback has found that feedback does not always improve performance. In some cases, it can make things worse. That makes sense when you look at how many reviews are vague, too focused on weakness, or aimed at the person's character instead of their behaviour.

You need to be more proactive

That sounds like feedback, but it gives the person nothing useful to do.

A better review answers three questions:

• What is working?
• What is getting in the way?
• What changes next?

The real outcome is a smaller blind spot

One useful way to think about performance reviews is the Johari Window. In plain English, a good review should reduce blind spots.

For the employee, that means understanding how they are actually seen. Not in a vague "you are doing well" way, but in specific behaviours. What should they double down on? What is creating friction? What are the two or three things that would make the biggest difference next quarter?

For the company, it means seeing the gap between expectation and reality. Is this person unclear on priorities? Are they overloaded? Are they quietly burning out? Are they in the wrong role, or just missing support?

This is where many reviews fail. They stay polite. They avoid the useful discomfort. The manager says, "You're doing great, keep it up," and the employee leaves with no new information.

That is not a performance review. That is a compliment with a calendar invite.

A simple 4-step performance review process

A useful performance review does not need to be complex. For most teams, 60 to 90 minutes is enough if the structure is clear.

1. Start with self-review

The employee should speak first and do most of the talking at the start.

Ask simple questions:

• What worked well?
• What was hard?
• What did you learn?
• Where do you see growth opportunities?
• What support would help you do better work?

This gives the manager two things: the employee's view of reality and the gaps in self-awareness.

Sometimes the employee already knows the issue. Sometimes they are focused on the wrong thing. Both are useful.

2. Give feedback with facts, not labels

Manager feedback should start with strengths. Not fake praise. Real examples.

What did this person do well? What impact did it have? What should they keep doing?

Then move to growth areas. Keep it narrow. Two or three areas are enough.

The rule is simple: every improvement area needs facts and examples.

Bad feedback: "You are not proactive enough."

Better feedback: "In the last three sprints, blockers were raised late. In two cases, the team only found out during the review meeting. I need you to flag risks earlier, even when the fix is not clear yet."

That gives the person something real to work with.

Feedback should describe behaviour and effect, not personality.

3. Discuss where your views match and clash

This is the part most managers rush.

After self-review and manager feedback, compare both views.

• Where do we agree?
• Where do we see things differently?
• What surprised you?
• What feels unfair or unclear?
• What are we not saying out loud?

This part can be uncomfortable, but it is often the most valuable. Misalignment is where the blind spot lives.

Maybe the manager thinks the person lacks ownership. The person thinks they have no decision rights. Maybe the employee thinks they are protecting quality. The manager sees missed deadlines.

The point is not to win the argument. The point is to understand the gap well enough to change the next step.

4. End with feed-forward, not a post-mortem

A review should not end with "here is what went wrong."

It should end with "here is what we do next."

Feed-forward feedback is about the next quarter, not the last one. It turns the review from a post-mortem into a plan.

A good plan should be short:

• What strengths will we use more?
• What two or three behaviours will change?
• What support is needed?
• What will the manager unblock?
• When will we check progress?

Over the next six weeks, you will post a short project update every Friday by 3pm. It should include progress, risks, and asks. I will help remove blockers within 24 hours where possible. We will review how this is working in our next 1:1.

That is much better than "communicate more."

Managers need to name the conversation clearly

If the review will be tough, say so.

This might be a hard conversation, but the point is to improve the system and the work, not punish you.

That sentence does two useful things. It lowers defensiveness, and it makes the standard clear.

Managers also need to know which role they are playing.

• As a coach, you help the person find their own insight.
• As a mentor, you share patterns and options.
• As a manager, you set priorities, boundaries, and decisions.

You can use all three roles in one review. But do not blur them.

If something is not optional, do not package it as coaching. Say it clearly.

HR should protect the rhythm, not own the whole thing

In small companies, HR or People Ops often tries to make reviews better by adding more forms. That rarely solves the real problem.

HR's job is to protect the quality and rhythm of the process. Managers still need to manage.

That means HR can help with:

• Clear templates.
• Review timing.
• Manager training.
• Calibration.
• Follow-up checks.
• Fairness across teams.

But the real conversation belongs between the manager and the employee.

The follow-up is where the review becomes real

Most performance reviews fail after the meeting.

The notes sit in Notion, Google Docs, or an HR system. Nothing changes in the calendar. No support is added. No blocker is removed. No check-in is booked.

Then everyone is surprised when the same issue appears again.

Minimum viable follow-through is simple:

• Send a short written summary after the review. Include goals, two or three actions, owners, and timelines.
• Make ownership clear. The employee owns their plan. The manager owns resources, clarity, and unblockers.
• Book a check-in within 4 to 8 weeks. Ask what moved, what got stuck, and what needs adjusting.

This is the part that turns a performance review process into actual management.

Because if nothing changes in the calendar, nothing changes in reality.