The Room Test

Twice a year your name comes up in a meeting you will never attend. Your manager is in that room. You are not. Whatever case exists for you at that moment is the whole case, and you cannot add to it, clarify it, or defend it.

I have sat in that room, including as an Amazon bar raiser. What I noticed over the years is that most people who stall are not failing at the work. They are failing a test nobody told them was being administered, and they find out months later in a sentence like "we could not determine individual contribution."

This is that test, written down. Take it on yourself, out loud, against a specific piece of work you are proud of. Answer honestly rather than generously. Being generous with yourself here is exactly what the room will not be.

You are not being judged on your work. You are being judged on what survives the retelling of your work by someone else, to strangers, in ten minutes.

How to take it

Pick one project. Not your year, one project. For each of the seven questions, answer yes only if you could produce the evidence right now, in the next ten minutes, without asking anyone. "My manager knows" is not evidence. "I could find it" is not evidence. Score one point per honest yes.

1. What did they actually do?

What the room is really asking: strip out the team, the org, and the quarter. What is the sentence that names this person specifically?

This is the question that kills the most cases, and almost nobody sees it coming. Strong engineers write in "we" because that is how good engineers talk. Then a stranger reads it and cannot tell which part was you. The room does not conclude you did nothing. It concludes it cannot tell, which lands in the same place.

Weak: "We rebuilt the ingestion pipeline and cut latency significantly."
Strong: "I found the fan-out that caused the tail latency, argued for the queue rewrite against two people who wanted to add caching, and owned the migration."

Point yourself one if you can state your contribution in one sentence that would still be true if every teammate read it.

2. Is this the next level, or more of the current one?

What the room is really asking: is this a bigger version of what they already do, or a different kind of work?

Volume is the most common failed argument at every level. Twelve of the same project is not the next level, it is a productive year at your current one. The levels above senior are not defined by more output, they are defined by output that does not route through your own keyboard.

Weak: "Delivered six major features, more than anyone on the team."
Strong: "Wrote the design that three teams then built against, which is why the six features were possible."

Point yourself one if you can name the thing you did that someone at your current level would not have been asked to do.

3. Who else saw it?

What the room is really asking: does anyone here, other than this person's manager, have an opinion about them?

Promotions are decided by people who do not manage you. If the first time a peer leader hears your name is when it comes up for decision, your manager is arguing alone, and a lone advocate reads as a favour rather than a consensus. This is not politics. It is the difference between one person's judgment and several people's.

Weak: nobody outside your team could pick your work out of a lineup.
Strong: two or three leaders outside your reporting line have seen you make a call, in a design review, an incident, or a planning fight.

Point yourself one if you can name the people, not the meetings.

4. Would it have happened without them?

What the room is really asking: is this ownership or participation?

Being on the project is not the same as being the reason it worked. The counterfactual is brutal and it is the fastest way to sort a room's opinion. If the project would have shipped roughly on time and roughly as well without you, you were staffed on it. That is fine. It is not a promotion.

Weak: "I was the main contributor on the migration."
Strong: "The migration was cancelled twice before I scoped it into three phases. That is what made it approvable."

Point yourself one if you can name what would have gone differently, specifically, in your absence.

5. What did it cost, and did they choose it?

What the room is really asking: did this person exercise judgment, or execute a decision someone else made?

Every real piece of engineering has a bill: something slower, something deferred, something someone was unhappy about. A case with no tradeoffs in it reads as either a small problem or a borrowed decision. Naming what you gave up is the single cheapest way to sound senior, and it is the part people edit out because it feels like admitting weakness. It is the opposite.

Weak: "Delivered on time with no issues."
Strong: "I cut the admin tooling to hold the date, took the support burden onto the team for a quarter, and we paid it back in Q3."

Point yourself one if you can name a real cost you chose, and why.

6. Would the notes survive someone disagreeing?

What the room is really asking: is there anything here a skeptic could check?

A case built from adjectives cannot be defended, because there is nothing in it to defend. "Strong technical leader" is not a claim, it is a mood. The moment one person in the room pushes back, an impression collapses and evidence does not. This is the same failure that ruins hiring debriefs, and it ruins promo packets for exactly the same reason.

Weak: "Widely respected, raises the bar for the team."
Strong: "Rewrote the on-call runbook after the March incident. Paging volume for that service went from a nightly occurrence to roughly none, and two other teams copied it."

Point yourself one if every claim in your case has something underneath it that a stranger could go and verify.

7. What happens to the team if they leave?

What the room is really asking: is this person a multiplier or a dependency?

This one is a trap in both directions, and it catches the hardest workers. If nothing breaks when you leave, you were replaceable. If everything breaks when you leave, you were a bottleneck, and a bottleneck is not a promotion case, it is a risk. The answer the room is listening for is that the team got permanently better and does not need you to stay that way.

Weak: "I am the only one who understands the payments service."
Strong: "Three people can now run a payments migration end to end. A year ago it was only me."

Point yourself one if you made someone else more capable in a way that outlasts you.

What your score means

Add up the honest yeses.

6 or 7. You do not have a promotion problem, you have a timing problem. Your case is real and it is legible. Write it down now, while the evidence is fresh, and make sure your manager has it in their hands well before the cycle rather than during it.

4 or 5. You have done the work and you have not built the case. This is the most common score among strong engineers, and it is the good news, because it is a documentation and visibility problem and those are fixable inside one quarter. Start with whichever questions you scored zero on. They are usually 1, 3, and 6.

0 to 3. This is not a writing problem, and no amount of rewording will move it. The work you are being given is not the work the next level is made of. That takes quarters to change, not weeks, and it starts with a direct conversation about scope rather than a better promo doc. Uncomfortable, and much better to know now than in eleven months.

One thing worth saying plainly: a low score is not a verdict on you. It is a verdict on the case, and the case is the part you control.

If you scored low and want a second opinion

Scoring your own case is hard, because the gaps you cannot see are the whole problem. That is what 1:1 coaching is for: I read the actual case, tell you which of the seven it fails and why, and we build the plan to close it. Book a 30-minute call and bring one project.

If you would rather work through it yourself first, The Top Engineer Method is this test expanded into the full system, and the essays are free.