“I don’t know if I’m going to get promoted or fired.”

That’s something engineers have told me more than once.

They’d walk into a performance review genuinely unsure whether the outcome would be good news or bad news.

Same company.

Same manager.

Same work.

From their point of view, it felt like a coin toss.

And honestly? I get why.

Most companies are still terrible at measuring engineering performance. And the larger or faster-growing the team, the worse it gets.

Engineers do their work. They ship. They help others. They solve hard problems. Then they wait.

They wait for a performance cycle to tell them whether all of that meant:

• promotion

• nothing

• or a problem

That uncertainty is the real issue.

Engineering productivity is hard. Pretending it’s easy makes it worse.

Engineering is not sales.

Sales is operational.

Engineering is creative.

In sales, you can usually measure outcomes with a single number. Revenue closed. Quota hit.

Engineering doesn’t work like that.

Engineering is about going from zero to one—building something that didn’t exist before. And that makes productivity much harder to benchmark.

This is why so many productivity dashboards fail.

They create the illusion of certainty.

They show numbers.

But they don’t show meaning.

And once numbers are used as weapons instead of signals, trust dies.

The mistake: treating productivity metrics like universal truths

Most leaders want simple answers.

They want productivity metrics that behave like physics:

• a clean score

• a ranking

• a red / yellow / green status

But engineering doesn’t work like that.

Because engineering is full of trade-offs:

• speed vs quality

• short-term vs long-term

• shipping vs stabilizing

• delivering vs refactoring

When someone asks, “What are the best engineering productivity metrics?”

The most honest answer is:

It depends.

Annoying—but true.

Measuring productivity creates fear when intent isn’t clear

The moment leaders say:

“We’re going to start measuring productivity.”

Engineers get nervous.

They’ve seen this movie before. It starts as “visibility.” Then becomes “control.” Then becomes “surveillance.”

And once that happens, you stop getting truth. You get politics.

If measurement is going to work, there need to be shared rules.

Engineers need to know:

• what’s being measured

• why it exists

• how it will be used

• what it will not be used for

Without trust, measurement is pointless.

The real problem isn’t people. It’s weak signals.

When performance reviews feel like coin tosses, it’s rarely because managers don’t care.

It’s because the system gives them weak signals.

They don’t have a clear line of sight into:

• what’s blocking delivery

• where quality is slipping

• where teams are overloaded

• who’s enabling others

• where work quietly disappears

Weak signals lead to weak decisions.

And weak decisions are what make performance feel arbitrary.

Final thought

If engineers are walking into reviews thinking:

“I don’t know if I’m going to get promoted or fired.”

Your system is broken.

Not your people.

In the next post, we’ll look at what high performance in engineering actually looks like—and why output alone completely misses the point.