What Good Looks Like: Defining Performance Before You Try to Measure It

Ask a leader to name their best performer and they will answer immediately. Ask them to describe what good looks like in that person's role, without mentioning hours, effort, or availability, and watch what happens.

The pause is the whole problem.

Last month I wrote about how organizations default to measuring time because value was never defined. The response I heard most was a fair one. Fine, so how do I define it?

Here is the answer, and it is smaller than people expect. Four questions, one role at a time.

Question One: What is this role here to produce?

Not the tasks. The result the tasks exist to create.

A job description is a list of responsibilities, and responsibilities are activities. Run the status calls. Send the reports. Answer the requests. Keep the account organized. Every one of those is on the list because it is supposed to produce something the business cares about, but the something almost never gets written down.

An account manager's real job is not running calls. It is a client who stays, grows, and refers. The calls are the activity. The renewing client is the point.

This is where most definitions stop before they start. If you can only describe the tasks, you have described the job, not the performance.

Question Two: What does good look like when it is happening?

Describe it the way you would describe it to the person replacing you. In behavior, not adjectives.

Not "proactive." Instead: the client hears about a risk from us before they discover it themselves. Not "strong communicator." Instead: when this person leaves a meeting, everyone in it can state the decision and who owns the next step.

The test is simple. Could two different managers read your description and recognize the same behavior? If not, you have written an impression.

Question Three: What is the signal you would watch?

One or two things. Not twelve.

Here is the distinction that does the heavy lifting. An activity measure tells you what someone did. A signal tells you whether it worked.

Calls made is an activity. Qualified second meetings is a signal. Tickets closed is an activity. Tickets that stayed closed is a signal. Reports delivered on time is an activity. Clients who renew without a price conversation is a signal.

Signals are harder to get. They are slower, noisier, and they often require waiting. Activity measures are fast, clean, and immediately available, which is exactly why they win by default in almost every organization.

Choose the signal anyway. You will get more of whatever you measure, and you would rather get the thing that worked than the thing that happened.

Question Four: What is the trap?

Every role has a convenient number sitting right there, and it is usually the wrong one.

Naming it out loud is protective. Once a team has said "the trap in this role is utilization" or "the trap here is response time," they stop drifting toward it without noticing. And they stop treating it as evidence when it shows up in a review.

There is a second reason to name the trap. Once people know what you are measuring, the measure starts shaping behavior. That is the point of measuring, and it is also how good metrics go bad. Put average call time on the board and calls get shorter, not better. Put tickets closed on the board and the same problem gets closed twice. None of that is dishonesty. It is people responding rationally to what you told them counts.

So run every proposed measure through one sentence. If someone wanted to make this number look good without doing the real work, how would they do it? If the answer arrives in under ten seconds, pair it with a second measure that keeps it honest. Usually that means one measure of volume and one of quality.

Then Put A Date On It

A definition without a timeline is still just a description. The last step is turning the outcome into something time-bound.

Write what the outcome looks like at three months, six months, and twelve months, specific enough that on the date, two reasonable people would agree on whether it happened. Three months is proof of motion. Six months is early signal. Twelve months is the outcome itself.

That structure does something quietly important. It gives you three moments to have an honest conversation instead of one, and each of them is early enough to still do something about what you find.

Why This Is Worth Twenty Minutes

Because the alternative is the conversation everyone dreads.

When the criteria live in a manager's head instead of on paper, the review becomes the first time anyone describes the standard. That is not a review. That is a surprise, and people do not absorb surprises, they defend against them. It is also where the quiet cynicism about performance reviews comes from. Not from being assessed. From being assessed against something nobody would show them.

Do the defining months before the conversation and the review gets boring, which is exactly what a good one should be. No new information. Nothing to brace for. Just two people looking at a standard they both already knew.

Take one role this week. Write down the four answers and the three dates. One page.

That page will tell you more about who is creating value on your team than any rating scale ever could.

If you want help turning your team's roles into clear, measurable definitions of performance, that is the work I do. Let's have a conversation.

Find the real problem. Do the real work. Build what lasts.