How to ask for a promotion: the scripts, the timing, and the mistakes that derail it

Two things quietly kill promotions: assuming the right people see your work, and waiting for your manager to bring it up.

This article covers the groundwork to do before you ask, when to bring it up, the exact words to use, and the mistakes that get strong performers passed over. Start months before you want the promotion, not weeks, because you need time to learn where you stand, make a plan with your manager, and then let the work play out where people can see the results.

The advice here comes from the deciding side of the table: Ethan Evans spent 15 years as an Amazon VP running promotion processes, and Deb Liu, CEO of Ancestry, has coached more than a thousand people through the ask.

The Magic Loop: build your case before you ask

Evans' Magic Loop is a five-step cycle he says has worked "for interns and vice presidents." It is deceptively simple and most people never get past step one.

Step four is where the loop pays off. After you have delivered for your manager, you say the goal plainly:

Deb Liu sees the cost of skipping that sentence constantly. Running mentorship circles, she tells the story of a woman stuck waiting (32:41 in her Lenny's episode): "You know what, I haven't asked. I was just kind of waiting for him to offer it." After coaching more than a thousand people who felt stuck, her first question is always the same, and the answer is usually the same too (48:18): "You'd be surprised at how many people said no, or how many people said yes and when I asked them how they asked, they didn't really ask." Managers are not mind readers, and hints do not count as asking.

Do the next level's job first

On My First Million, Shaan Puri says most people have the order backwards:

His concrete version: be a plan maker, not a plan taker. Write your own version of the annual plan before anyone asks you to. He adds two force multipliers from his own career. First, pick your manager deliberately (18:06): your promotion odds rise and fall with your manager's clout, so choose someone with real influence, and consider switching if yours has none. Second, the 5 percent rule he learned from his Twitch cofounder (19:46): only take projects that could visibly move company-level numbers by at least 5 percent. A small project done perfectly will not get you promoted. A visible one might.

When to ask: earlier than feels polite, smaller than feels satisfying

Evans gets asked how early is too early. His answer reframes what the first ask should be:

You do not open with "promote me." You deliver some value, then ask for the expectations: what does the next level look like here, and what would you need to see from me? That conversation can happen months or years before the real ask, and it turns your manager into a collaborator instead of someone hearing a demand.

Know where you stand, and say so

The expectations conversation gives you the benchmark. What you do next is keep your own written read of how your work maps to it, project by project, so every career conversation starts from your opinion instead of an open question. Evans builds this directly into the Magic Loop:

The difference matters. "How am I doing?" invites reassurance. "I think I'm doing that. Am I?" forces a specific answer against a benchmark you both agreed on, and when the answer is no, it usually comes with the exact gap you need to close. When you write the mapping, favor evidence over responsibilities: results, expanded scope, decisions you got right, and work that made other people better.

Deb Liu's version of the same discipline is the self-review. She keeps meeting strong operators who avoid writing about their own work because it feels like bragging:

She has also seen what happens without that mapping. Earlier in the same episode (34:36) she tells the story of a brilliant PM on her team whose name came up at every calibration with the same question: what does she do? When a room of managers discusses your promotion, you're not there to explain your work. The written record, in self-reviews, in your 1:1 doc, in the case your manager carries, does the explaining for you.

Build your coalition

Your manager is one voice in the decision. Peers and skip-levels weigh in, formally in calibration and informally whenever your name comes up. Carla Harris, vice chairman at Morgan Stanley, has spent decades on the deciding side of those conversations, and she is blunt about what that means:

Her playbook for getting those advocates is specific (7:15 in the same episode). Study your organization for two weeks and note who has a seat at the decision-making table; if you cannot tell, ask someone. Make sure those people have visibility into your work. Then pick two or three of them and build the relationship through what she calls frequency of touch: small, repeated, low-stakes interactions until they feel they know you. She describes what it looks like from the sponsor's chair:

This is also why the "who else weighs in" question from the callout above matters. Ethan Evans lived the other side of it at Amazon: in his 15-year retrospective he talks about watching peers get promoted to director and VP before him, "and it sucked." His advice from that stretch was to work near stars, because you become the people you spend time with, and their advocacy adds up. Deb Liu's calibration story earlier makes the same point in reverse: brilliant work with no advocates in the discussion gets reduced to "what does she do?"

The company has to need it too

Being ready is only half of a promotion. The other half is that the company needs someone at that level, and it may not. If you want to move into management, someone has to need managing. A team that is not growing may have no room for another manager, and no amount of performance changes that.

What creates those roles is growth. Zhuo says her own path into management happened "because Facebook was always scaling," and that at a growing company there is always more opportunity because somebody has to do the new work (7:40 in the same episode). Flat headcount works the other way: a manager slot opens only when the current manager leaves.

Nikhyl Singhal, VP of Product at Meta, explains why those slots have been getting scarcer across tech, and it has nothing to do with any individual's performance:

So put the question about need into the same early conversation where you ask what the next level looks like: does the team have room for another person at that level, and if not, what would create it? An honest "we don't have the need right now" is useful information, and it leaves you three real options. Find the need on another team. Grow your scope where you are and let the title come later. Or, as Zhuo puts it, move to an environment that is growing fast enough to need you at that level, because you can do everything right and still be waiting on a role that does not exist where you are.

The ask itself: the script

When the evidence exists, Evans gives the exact sentence:

He stresses the pronoun: "we," as in how could we go about that. The claim is evidence ("delivering value equivalent to this level"), the goal is explicit, and the framing recruits your manager onto your side of the table.

One sharpening from the Mochary Method curriculum, which coaches CEOs on how they set pay: companies anchor compensation to the market rate for a role and level. Effort and personal need never enter the formula. So the strongest version of the ask is scope-based: Say "my scope now matches level X" and you're stating a fact about the job you're already doing. "I have been working hard" and "I need more" read as requests for a favor. Same conversation, different category.

Then put it in writing

Whatever the two of you agree, send a short recap afterward so you both have the same record:

The mistakes that derail it

Assuming good work is visible

Adam Goyette was a senior marketer when an Ivy League MBA intern got promoted over him. His diagnosis, years later on Delivering Value:

It was not enough. The founders making the decision could not see his work. His fix (17:03): "I started celebrating my wins, like, publicly." Internal show-and-tell of wins and experiments, translated into revenue terms the decision makers cared about. Promotion decisions happen when you are not there, and the people making them have to know your work, not just your boss.

Waiting to be offered

Liu's stuck-mentee story above is the pattern: treating the promotion as something bestowed rather than something asked for. Her structural fix from the same episode (29:10): write out where you want to be in two and five years, write the pre-mortem for how your career drifts if nothing changes, and work backwards. She calls it PMing your career like a product. Every project you take either moves you toward the destination or it does not, and you cannot evaluate that without writing the destination down.

Breaking the loop when you get passed over

Sometimes you ask and the answer is no, or worse, someone else gets it. Sarah Ellis, co-host of the FT's Working It "squiggly careers" episode, has a first step most people skip: name the emotions you are feeling, but not to somebody you work with (4:05 in the episode, audio only). Vent outside the building, with a friend or a coach, then come back and ask the colder questions: was the case invisible, was the timing wrong, or is the ladder itself the problem? Her second point (6:50): if progression only means promotion, flattening orgs will break you. Scope, skills, and lateral moves are progression too, and sometimes they are the faster route to the title.

For the specific failure patterns in product careers, Nikhyl Singhal's Skip episode on the promotion mistakes that derail PM careers is a strong companion to this piece.

The playbook on one page