Running a Performance Review That Isn't a Waste of Time

Why this matters

The annual review most shops run is a ritual nobody benefits from: the owner scrambles to remember a year the night before, the tech sits through a graded report card, both leave relieved it is over, and nothing changes. That is not a review, it is paperwork. A real performance review is one of the few times a year you and a tech step off the treadmill and answer two questions that actually shape the next year: how did the last stretch go, and what are we building toward. Done right it drives retention, development, and honest accountability. Done as theater it quietly teaches your best people that you are not paying attention.

The one rule that fixes most reviews: no surprises

If anything in the review is news to the tech, you failed during the year, not in the meeting. A review is a summary and a look forward, never the debut of a problem. The tech who first hears about a recurring issue in their annual review is right to feel ambushed, and the message they take is that you sat on it for months. Correct in the moment, all year. The review then becomes a place to connect the dots, not to drop a bomb. If you find yourself saving up grievances for the review, that is the habit to fix first.

Keep notes all year so it is not a memory contest

The reason reviews collapse into recency bias, judging the whole year by the last three weeks, is that nobody wrote anything down. Keep a running note per person: a standout job, a callback that mattered, a good customer comment, a missed commitment. It takes seconds a week and it turns review prep from a stressful reconstruction into a five-minute read-through. It also lets you say "on this specific job in the spring" instead of a vague impression, which is the difference between feedback a tech trusts and feedback they discount.

Split the meeting cleanly: look back, then look forward

A review that only grades the past is a scorecard, and scorecards do not change behavior. Structure the time in two halves.

• Look back: how the year actually went against the standards you set. Name the wins first and specifically, most owners rush past these, and the strong stuff is exactly what you want more of. Then the gaps, tied to real examples, framed as things to close, not sins to confess.
• Look forward: what the next stretch is for. One or two goals that matter, a skill to build, a certification to chase, a step toward where they want to go. This half is what a good tech shows up for, and it is the half most reviews never reach.

Make it two-way, or it is just you talking

A review where only the manager talks tells you nothing you did not already know. Ask the tech to self-assess before you weigh in: "Where do you think you were strong this year, and where do you want to get better?" People are often harder on themselves than you would be, which lets you build them up instead of tearing down. Then ask the questions that surface what you cannot see: What is getting in your way? What would make the job better? What do you want to be doing a year from now? The answers are early warning on a tech who is drifting toward the door, and you would rather hear it in a review than in a resignation.

Keep pay out of the room, mostly

This one is counterintuitive. If the review is where a raise gets decided live, the tech spends the whole meeting managing the money and cannot hear a word of the development conversation. Separate the two: let the review inform pay, but deliver a pay decision as its own conversation, before or after, not during. A tech braced for the number is not absorbing feedback. There are shops where a single combined meeting is unavoidable, and if so, handle the pay decision first and cleanly so the rest of the conversation is not held hostage to it. See related: The Review Cycle That Should Drive a Pay Change.

End with something written and something owned

Close the same way every time: a short written summary both of you keep, and a small number of forward commitments with dates. Not a ten-point plan, one or two things that will actually get worked on before the next check-in. Then put a mid-year check on the calendar so the goals are not filed and forgotten. A review that ends in a document nobody looks at again is back to theater. A review that ends in two owned commitments and a date to revisit them is a management tool.

The tell that it worked

You will know the review did its job by one sign: something is different afterward. The tech is working on a named skill, a gap is closing, or you understand something about them you did not before. If you cannot point to a single thing that changed because of the meeting, you held a ceremony, not a review, and it is worth rebuilding before you run the next round.

References

• Society for Human Resource Management (SHRM), conducting effective performance reviews
• U.S. Small Business Administration (SBA), employee evaluation and development practices
• See related: The Performance Conversation That Changes Behavior; The Review Cycle That Should Drive a Pay Change