TL;DR
- The dread has a structural cause: mid-level, someone senior stress-tested your calls before they shipped. Becoming the ceiling deleted that check.
- Split the doubt before treating it. Fraud fear, a real skill gap, missing recognition, recalibration at new altitude, and overselling need opposite responses.
- AI can hand you information, but a model tuned to agree with you can't calibrate your judgment. It has never watched your exact call fail.
- Rebuild the loop on purpose: one outside operator who has held your seat, with standing permission to argue with you before you commit.
- Feed it formed decisions. Bring the call, your reasoning, and what would change your mind, and run a weekly self-review in between.
The first call you can't run past anyone
Here is the moment this post is about. A pricing change, a key hire, the plan for the quarter. You finish the doc and reach for the person you'd normally send it to with "poke holes in this?", and the chair is empty. You are now the most senior person on this. Whatever you decide ships.
The disorientation is newness, and from inside the seat newness reads as inadequacy. Charity Majors named the seat the trap of the premature senior, the person who became the most senior in the room before anyone finished teaching them how to hold it.
The internet's advice is a shelf of posts telling you to stop waiting for a mentor and learn from your peers. Every one of them stops at the same line, the moment you commit an unchecked call and nothing downstream catches it. That moment is the entire problem, and it deserves better than a pep talk. The refusal to name it is why those posts feel thin at exactly the moment you need them.
So let's name what vanished.
the loop you lost
Mid-level, last year
Promoted, this year
Rebuilt, on purpose
The middle node was doing invisible work, and promotion deleted it. The third row is the only version you control.
The day you became the ceiling, the check disappeared
Rewind to mid-level. Every consequential call you made passed through someone senior on its way out. A boss trimmed the budget, a lead flagged the risk you missed, a director asked one question that saved the launch. You experienced that as friction. It was an instrument, the thing that told you whether your judgment was good or just lucky.
Promotion deleted the instrument. You kept making calls, you stopped getting readings, and the doubt that filled the silence got a market name, impostor syndrome. Mostly it's misdiagnosed. Your competence didn't change in the elevator ride up. Your feedback did.
And this stopped being a rare career accident. The apprenticeship rung where a senior checked a junior's work is the rung AI ate. Stanford's read of ADP payroll data puts early-career employment in the most AI-exposed jobs down 13% since generative AI arrived, entry-level software down about 20%. Fewer people get checked on the way up, and more land in the ceiling seat early.
"Suddenly senior with nobody above you" went from a rare career accident to the default setup. You aren't imagining the missing safety net, the industry removed it.
The stand-in most people reach for makes it worse. 49% of young workers now take work questions to ChatGPT over their manager, and founders prompt it to act as a brutally honest advisor. For information, fine. For calibration it hands you a mirror, not a check, because a model tuned to agree with you has never watched your exact call blow up in production.
Agreeable is the one thing a check can't afford to be.
Name which doubt you have before you treat it
The reframe is a diagnosis, so run it on yourself first. "I feel like a fraud," "I've genuinely never done this before," and "I know I'm right but I can't back myself" are three different problems that want opposite treatments, and stacking them into one blob called impostor syndrome is why the blob never moves.
Good mentors refuse to treat the label. The first half of a good session is questions, because the best thing a mentor does is tell you you're solving the wrong problem. One version of the split I've watched work in practice breaks the umbrella into five distinct doubts.
which doubt is it
1 · The fraud fear
“They'll find out I'm winging it.”
Audit the claims you have made against what you have done. Almost nobody finds an actual lie.
2 · The real gap
“I've genuinely never done this.”
Correct, and it is a skill gap in a new seat. Learn it like one.
3 · The missing signal
“I'm doing fine and nobody says so.”
A recognition gap. Go ask for the reading you stopped getting.
4 · The altitude effect
“The more I see, the less sure I get.”
Recalibration. Exposure widened and certainty dipped. This one arrives with every promotion.
5 · The oversell
“I claimed more than I've done.”
The one honest gap. Close it fast, before someone else measures it.
Five problems under one label, each with a different fix. The newly senior usually hold the fourth, and it responds to calibration.
Sort yourself honestly and most of the menu disappears. Whether your case is a fraud feeling or a genuine skill gap has its own longer treatment. The newly senior operator usually lands on the fourth doubt, the one that arrives with altitude, because for the first time nobody is absorbing your errors before they count. If the sentence in your head is closer to "I don't trust my own judgment right now," that's the fourth doubt talking, and it's the one that deserves the most attention.
Rebuild the reference point, with someone who has held the seat
A loop needs an external node, and the kind matters. A coach asks how the decision makes you feel. A peer shares your blind spots. What the loop wants is an operator a level up in experience who has sat in your seat, used as a sparring partner for validation, with standing permission to argue with you before you commit the org.
This is the part I can't be neutral about, because I built a company out of it. In 2018 I was running marketing with nobody above me, and I started paying freelancers on Upwork to argue with my plans before I shipped them.
That habit became GrowthMentor. The loop was always the product.
The stall that keeps people from doing it is "I should get more experience first." Run the math instead. An outside hire spends six months learning your company. You'll spend six months learning the function. The net position is identical, so the waiting buys you nothing. And unprompted honesty gets rarer as you rise, feedback thins near the top, which is why the current consensus fix is a personal board of advisors rather than one mentor above you in the org chart.
Two reframes make the seat lighter while you build the loop. The senior job was never being the most technically superior person in the room, it's creating the conditions where strong people do their best work, measured by team output. And you don't have to present as a guru, position as someone one step ahead, recently down the same road. Both are true, and both are easier to believe once someone qualified has checked your reasoning.
The sharpest version of this I know is a CEO with no senior marketer in the building sending a new growth lead off with "get a second opinion on the recovery plan and flag any gaps." That is a leader rebuilding the missing check on purpose. Same instinct as a first-time CEO borrowing judgment for the layoff call, the seats where the empty chair bites hardest.
Borrow a senior who has made your call before
One membership, unlimited 1:1 calls. Pick an operator who has held your seat, hand them your reasoning, and get it argued with before you commit.
Bring a call, not a blank page
The loop only works if you feed it a formed decision. Walk in with "what should I do about pricing?" and you get a lecture anyone could give. Walk in with "here is the call I'm about to make, here is my reasoning, tell me where it breaks" and you get calibration, the crack found before you commit, plus a read on which of your instincts are local artifacts of the one company you learned in.
what the ask buys
Where the weight sits
A formed call
What floats up
A blank page
Load the pan with a formed decision and the session pushes on your reasoning. The blank page floats, and it buys a lecture anyone could give.
And triangulate. Any single voice is one input, the operator's included, the AI's included. A second opinion you can trust is a system you assemble, one input at a time. The good operators model this themselves. Midway through a session they'll tell you to get a second opinion, sometimes referring you to another mentor by name, and they'll admit where their own read runs out.
You already did the thinking. The loop's job is the crack.
Calibrate in the weeks between
You can't book a session before every decision, so the loop has to run when nobody's watching too. The tell of a real senior was written down years ago, and it isn't having the answer. It's knowing how sure to be. That is a skill you train against your own past self, on a schedule, without a phantom senior and without the LinkedIn highlight reel. Mentors treat confidence at this altitude as a practice-volume problem rather than a knowledge gap, and volume is schedulable.
How-to guide
The self-review loop, weekly
The part every grow-without-a-mentor article skips. Five habits that replace the readings you stopped getting.
Keep a decision log
One line per consequential call. The decision, your confidence from one to ten, and what evidence would change your mind. A minute to write, and in three months it's an instrument panel.
Run a 45-minute Friday review
Reread the week's calls. Separate what you knew at decision time from what you learned after, and grade the reasoning, never the outcome. Good calls lose sometimes. Lucky ones win.
Benchmark against your own past self
Compare this quarter's judgment to yours six months ago, never to another operator's highlight reel. The past-you baseline is the only one with the same information and the same seat.
Log the wins
A running list of things that worked because of a call you made. Confidence at this altitude is a practice-volume problem, and evidence beats affirmation every week of the year.
Escalate the org-committing calls
Anything that moves money, people, or the roadmap gets the outside pressure-test before it ships. The self-review keeps you calibrated between those, it never replaces them.
Your one move this week
Take the decision sitting on your desk right now and give it one page. The call in one sentence. The reasoning underneath. What evidence would change your mind. Writing it is half the calibration, because most fog doesn't survive the second box. The whole page takes ten minutes, roughly the length of the email you were going to send instead.
the page you bring
One decision · one page
The call, in one sentence
We move the annual plan to $79 on March 1.
The reasoning
Support load per customer doubled this year. The two rivals we lose deals to are priced at $89 and $99. Churn held steady through the last increase.
What would change my mind
Trial-to-paid under 8% in the first month, and we roll it back.
Three boxes, ten minutes. The second kills most of the fog, and the third is the first thing a good pressure-test grabs.
Then put the page in front of one person who has held your seat, this week, before you commit it. The page is the loop's input. Running it once, on purpose, is the whole assignment.
What you're buying back
The payoff is more specific than feeling better. It's the shift from "am I doing this wrong?" to "I can keep going," permission to trust a judgment that has now been checked by someone qualified to check it. Most operators discover they already knew the call. What was missing was someone who has been the ceiling, saying it out loud before it shipped. Half of CEOs report the seat is lonely, and first-timers report it worst, which is exactly why the check has to be built instead of waited for.
When people ask what GrowthMentor is for, this is the honest answer. It turns the empty chair into a standing loop. You describe the seat you're in, in one sentence, and the matching puts vetted operators who have sat in it in front of you, each with a line on why they fit. One membership covers unlimited 1:1 calls with every one of them, so the pressure-test can become a habit instead of a splurge.
Take a look at these mentor profiles and edit & tune your challenge accordingly if you want to see different suggestions.
I’m the most senior marketer at my company and I need someone who has run the seat to pressure-test my Q3 plan
Edit & Tune Results
Craig has founded six companies, so the calls you can’t run past anyone are calls he has already made. He pressure-tests plans for operators who are the most senior person in the room.
Sergi is the most senior marketer at Red Points, the seat you are describing. He knows which parts of a quarterly plan get challenged, and he will challenge yours the same way.
Nick led product and growth teams at Shopify, where plans got run past him for years. He will find where the reasoning breaks before your board does.
Eight years on, there is still nobody above me. The scary calls still get pressure-tested every week, by people who have made them before me. The chair stayed empty. The loop runs anyway.
The mentors below have been the ceiling, repeat founders and Heads-of who now spend part of their week checking other people's calls.
Suggested mentors
Operators who have held the senior seat, on the roster now:
No one above you FAQ
Vetted mentors, every one included
Someone used to check your work.
Rebuild that, on purpose.
Browse operators who have held your seat, book a 1:1 call on-demand, and bring the decision on your desk. Membership is unlimited calls, every mentor included.
Talk to a mentorKeep reading
More from the GrowthMentor blog
Mentorship · Jul 01, 2026
The Best Thing a Mentor Does Is Tell You You're Solving the Wrong Problem
Foti Panagiotakopoulos
Leadership · May 27, 2026
First-time CEO, and the first thing on your desk is a layoff
Foti Panagiotakopoulos
Mentorship · May 19, 2026
How to tell impostor syndrome from an actual skill gap
Foti Panagiotakopoulos
Mentorship · Apr 28, 2026
When you need a second opinion you can actually trust, not just the one that agrees with you
Foti Panagiotakopoulos
