Why good people get "not this cycle"
Between November and January, most companies run their review and promotion cycles. Your manager takes your self-review into a calibration meeting with other managers who have never seen your work. They have a few minutes per person and a fixed budget of promotions.
In that room, a case lives or dies on one thing: can a skeptical stranger verify it quickly? "Helped with the migration" cannot be verified. "Led the billing migration across 3 teams, cutting failed renewals by 18%" can. Same work, different outcome.
Whatever the company and whatever the rubric, the discussion keeps coming back to some version of the same five questions. If your case answers them on the page, your manager does not have to improvise in the room.
1. "What changed because of you?"
This is the impact question. Committees are not grading effort or activity; they are looking for a before and an after. Lines that start with "worked on", "helped", or "was involved in" describe activity and invite this question immediately.
Invites the question
Worked on improving checkout reliability.
Answers it
Cut checkout errors by [X%] over [period] by fixing the top [number] failure modes, recovering about [orders or revenue] per month.
Fix: Rewrite each bullet to start with the result, then how you got there.
2. "How do we know?"
This is the evidence question. A claim without proof is an opinion, and opinions are easy to discount when budgets are tight. Proof can be a metric, a dashboard, a design doc, a launch post, or a named person who can vouch for it.
Invites the question
Significantly improved onboarding for new engineers.
Answers it
Rebuilt onboarding for [number] new hires; time to first merged PR dropped from [before] to [after] (team dashboard), confirmed by [manager or peer].
Fix: Next to every big claim, write where the proof lives. If there is none, find it now, not in the meeting.
3. "How big was it?"
This is the scope question, and it is what separates levels. The same accomplishment means different things for one team, five teams, or the whole company. Committees need the scale spelled out: users, teams, systems, revenue, duration.
Invites the question
Improved our CI pipeline.
Answers it
Cut CI time from [before] to [after] for [number] engineers across [teams], saving roughly [hours] of waiting per week.
Fix: Add who and how many to every outcome. Scale you leave implicit does not count.
4. "Was this you, or the team?"
This is the ownership question. Committees are wary of team results presented as individual ones, and equally of people who undersell their role with "we". The answer is to name your specific decision or contribution while giving the team its credit.
Invites the question
We migrated the payments service with zero downtime.
Answers it
I designed the dual-write approach that let the team migrate the payments service with zero downtime, and ran the cutover across [teams].
Fix: For each team win, add one sentence that starts with "I decided", "I proposed", or "I led".
5. "Is this the next level, or the current one done well?"
This is the alignment question, and it sinks more strong cases than any other. Excellent execution at your current level earns a great rating, not a promotion. The case has to show you already operate at the target level: broader scope, more ambiguity, direction set, others multiplied.
Invites the question
Shipped 4 major features this year on time.
Answers it
Identified [problem] as the biggest drag on [org] and set the plan [teams] followed, which [outcome]: the scope [target level] is expected to own.
Fix: Open your company's level rubric. Map each of your top three outcomes to a specific line in the target level, in its own words.
The 30-minute pre-review checklist
Before you submit, run through this once:
- Your top 3 outcomes each start with a result, not an activity
- Every big claim has a number, artifact, or person behind it
- Scale is explicit: who, how many, how long
- Each team win names your specific decision
- Each outcome maps to a line in the target-level rubric
- One honest growth area, with what you have already done about it
- No number you cannot defend if someone asks where it came from
We built a free tool that asks these questions for you
YouDeserveNow reads your accomplishments the way a skeptical committee would. Paste your draft and in about a minute you get a score across evidence, impact, scope, leadership, and level alignment, the three questions your case is most likely to get, each tied to the exact line that triggers it, and your weakest line rewritten.
It never invents numbers. Anything only you can supply stays as a highlighted [placeholder], so everything you submit is defensible. The audit is free with no sign-up, cases are encrypted and deleted after 30 days, and if you want the finished self-review, one-page case, meeting script, and objection responses, the Case Pack is $29.