What Your Manager Says About You When You're Not There

Your manager is sitting in a conference room right now. There are six other managers at the table, each with a stack of names. Your name comes up. Your manager has about 90 seconds.
What do they say?
Not what you hope they say. Not what they told you in your last one-on-one. What they actually say to a room full of people who have never read your code, never seen your design docs, and never attended your standups.
That answer — delivered in a calibration meeting you'll never attend — determines your rating, your compensation adjustment, and whether your promotion goes through or gets tabled for another six months. Most engineers never find out what was said. They just get the result.
The 90-Second Pitch You Never See
Calibration meetings run on a clock. There might be 40 people to discuss in a two-hour block. The math works out to about three minutes per person, and most of that is back-and-forth between managers, not presentation time. Your manager's actual window to make your case is closer to 60 or 90 seconds.
In that window, they need to land three things: what you did, why it mattered, and what level it maps to. If they can do that clearly, the room moves on and the rating sticks. If they can't, other managers start asking questions. Questions are not good. Questions mean the case didn't land. And once a case starts unraveling in the room, it's hard to put it back together.
Research from Harvard Business Review found that speaking order and verbal persuasiveness directly affect calibration outcomes. The most articulate manager doesn't always have the strongest performer. But they often get the strongest rating through.
This isn't fair. It's also not going to change next quarter.
What a Strong Case Sounds Like
When a manager walks into calibration prepared, the pitch sounds specific and compact. Something like:
"She led the migration of our payments service to the new API gateway. Reduced P99 latency by 40%, eliminated a class of on-call alerts that was generating three pages a week, and coordinated across three teams to get it shipped in one quarter. That's staff-level scope and execution."
That pitch has a specific project, quantified outcomes, cross-team coordination, and a level claim. The room can evaluate it. Other managers can nod or push back, but they're responding to something concrete.
Strong cases share a few patterns:
- One or two projects, not a laundry list. Managers who try to rattle off everything you did in six months lose the room. The narrative that works is a focused story about your most impactful work.
- Numbers. Latency improvements, revenue impact, incidents prevented, teams unblocked. Numbers cut through opinion.
- Scope language. "Cross-team," "org-level," "drove the decision" tell the room what level the work was at. Without these markers, the room defaults to "solid IC work" and moves on.
- A clear level claim. "That's senior-level impact" or "this is the kind of thing we expect at L6" gives other managers a frame. Without it, they apply their own, and their own is usually more conservative.
What a Weak Case Sounds Like
When a manager walks in without preparation, or without the raw material to prepare from, the pitch sounds different. It sounds like:
"He's a strong technical contributor. Very reliable. The team really depends on him."
That sentence says nothing. Every manager in the room can say the same about their own people. It doesn't name a project. It doesn't quantify anything. It doesn't claim a level. It just takes up 15 seconds and creates silence.
Weak cases have their own patterns:
- Adjectives instead of evidence. "Reliable," "strong," "solid," "great teammate." These words feel positive but carry no information in a calibration room. They're filler.
- No project specifics. If your manager can't name the project you led, can't explain what you built, and can't describe the outcome, they have nothing to defend when someone pushes back.
- Defensive hedging. "He did a lot of good work this half" followed by a long pause. The room reads that as uncertainty. And uncertainty in calibration means the rating moves down, not up.
- Character testimony instead of impact. "Everyone loves working with her" is a fine thing to say about a person. It's not a case for promotion. Calibration rooms evaluate scope and outcomes, not likability.
The batting analogy keeps coming up when managers talk about calibration. You're going up to bat on behalf of someone else. If you can't make contact, you both strike out.
Why Your Manager Might Have Nothing to Say
Here's the part that stings. Your manager might genuinely believe you're doing excellent work. They might have told you so in your one-on-one last week. They might plan to fight for you in calibration.
And then they get to the room and realize they can't articulate what you did.
This happens more often than most engineers think. Not because the manager doesn't care. Usually it's one of these:
They don't have the raw material. You never gave them a written summary of your biggest wins. You never quantified the outcomes. They know you "worked on the data pipeline" but can't explain what you changed, why it mattered, or what the before-and-after looked like. So they improvise.
Your work is hard to explain. You spent three months on a complex refactor that prevented a class of production bugs. The technical value was real. But explaining that to a room of managers who own different product areas, in 90 seconds, without visual aids, requires a level of translation your manager may not have prepared.
They're managing too many people. A manager with eight direct reports has to prepare eight cases. Some of those cases will get more attention than others. The engineers who made their work visible all year are easier to pitch. The ones who kept their heads down require the manager to reconstruct six months of work from memory. Memory is unreliable. Something always gets lost.
They never asked, and you never told. Some managers are proactive about collecting evidence for calibration. Many are not. If your manager never asked you for a summary of your impact, and you never volunteered one, the gap between what you did and what they can say about it grows wider every month.
Deb Liu, former VP at Meta, wrote that a manager's organizational level affects how their advocacy is perceived in calibration. When a senior leader advocates for someone, it reads as a fair assessment. When a frontline manager makes the same argument, it can read as partisan. Your manager is already fighting from a disadvantage. If they also don't have specifics, they're fighting unarmed.
The Difference Between Knowing and Guessing
There are two versions of the manager who walks into calibration on your behalf.
The manager who knows your work had it handed to them. You wrote a half-page summary of your top two accomplishments. You connected each one to the rubric. You quantified the impact. You named the stakeholders. When someone in the room asks, "What did she actually do on that project?" this manager has the answer ready. They're not thinking. They're reading from the case you built together.
The manager who has to guess is filling in the gaps in real time. They remember you worked on something related to the alerting system. They think the project landed well. They're not sure if you led it or contributed to it. When someone pushes back, they hesitate. The hesitation is the tell. The room notices.
The gap between these two managers is not about talent or how much they care about your career. It's about what information they had access to before the meeting started.
Engineers spend months building real things and then leave the hardest part of the promotion process to chance: whether their manager can find the right words at the right moment. That gap is closable.
How to Change What Gets Said
You can't be in the calibration room. But you can put words in your manager's mouth. Not manipulatively. Literally. Give them the sentences.
Write the pitch for them
Two weeks before review season, send your manager a short document. Not your full self-review. A half-page summary of your two or three most impactful projects, with outcomes quantified and scope made explicit. If you haven't built one yet, writing a promotion case document walks through the format that makes your manager's job easiest.
Frame it the way you'd want it repeated in calibration:
"Led the reliability overhaul for the checkout service. Reduced incident rate by 60% over two quarters. Coordinated across payments, platform, and SRE. Resulted in the team being removed from the weekly SEV review escalation list."
Your manager may not use your exact words. But they'll have something specific to work from instead of reconstructing your year from fragments.
Name the level explicitly
Don't leave level mapping to your manager's interpretation. If you believe your work was at the senior level, say so. "I think this maps to senior-level scope because it required cross-team coordination and I drove the technical decision-making." Your manager can agree, push back, or adjust. But you've started the conversation with a frame instead of leaving it open.
Make your work visible before calibration season
The engineers who are easiest to promote are the ones other managers already know. If another manager in the calibration room has seen your name on a cross-team project, attended a presentation you gave, or worked with you on an incident, your manager's pitch gets validated by a second voice. That's more powerful than any written evidence.
If your work is strong but the people making decisions don't know your name, you're invisible in the room that matters. Visibility is not about self-promotion. It's about making sure the people in the room have data beyond your manager's word alone.
Update your manager regularly, in writing
A weekly or biweekly update email is the lowest-friction way to ensure your manager always has material. Not a status report. A quick summary of what you shipped, what it unblocked, and any recognition you received. When calibration time comes, your manager has six months of receipts instead of six months of guessing.
Ask what they'll need
Some managers will tell you directly: "Send me your top wins before calibration." Others won't think to ask. Either way, the question "What do you need from me to make the strongest case in calibration?" puts the collaboration on the table. It tells your manager you understand the process, and it makes it easy for them to tell you what's missing. If you haven't had that conversation yet, how to have a career conversation with your manager covers exactly how to open it without it feeling forced.
The Conversation You're Not Having
Most engineers treat calibration like weather. It happens. They find out the results afterward. They either got promoted or they didn't.
But calibration is not weather. It's a structured process with identifiable inputs, and you control more of those inputs than you think. The single biggest input is what your manager says about you. And the single biggest factor in what they say is what you gave them to say.
If you've been passed over and the feedback was vague, or if your manager seemed supportive but nothing happened, the gap probably wasn't your performance. It was translation. Your work existed but it didn't survive the 90-second pitch.
That's fixable. Not by working harder. By writing it down, framing it clearly, and making sure the person who speaks for you actually has something to say.
Your manager shouldn't have to guess what you accomplished. CareerClimb helps you capture wins as they happen and build documented evidence for your promotion case, so when calibration comes, your manager has more than memory to work with.



