The probation period (испытательный срок) is the only stretch of your career when everyone expects you to ask questions, and almost nobody uses it. New hires spend the first three months being polite, invisible, and busy — then discover in month four that nobody can name a single thing they delivered.
Here is the uncomfortable part: a probation review is not an exam you pass by not failing. It is a decision made by people who are already overloaded and rely on memory. If your manager cannot recall a concrete result when the promotion conversation starts, the default answer is "let's revisit in six months." That answer is not personal — it is what happens when evidence is missing.
The plan below is built around one idea: you are not trying to look good, you are trying to make your results easy to remember and easy to defend. That is the whole mechanism — everything else is scheduling.
Why the first 90 days decide more than the next three years
The probation period is a strange contract. The company has committed to you, but only partially; you have committed to them, but can leave cheaply. Both sides are watching for the same thing — not skill, but predictability. Can this person be handed something on Monday and not need to be rescued on Friday?
This is why the person who delivers three boring, finished things beats the person who promises one exciting thing. Finished work becomes a data point other people can repeat to each other. Unfinished ambition becomes a story that only you remember.
There is also a timing asymmetry most people miss. By week six, your manager has already formed a rough verdict — usually privately, usually without telling you. Everything after that either confirms or contradicts it. So the work of weeks one to six is not "learning the job." It is giving the verdict something good to land on.
Week one: map the terrain before you impress anyone
The most common first-week failure is trying to contribute before understanding what "good" looks like here. Every company has an informal definition of a job well done — a template, a review habit, a person whose approval matters. None of it is in the onboarding document.
What to actually do in week one:
- Write down who decides what. Not the org chart — the real one. Who approves your work, who can veto it, who your manager asks before answering you.
- Ask your manager one question: "In 90 days, what would make you say this hire was a clear success?" Write the answer down verbatim. This single sentence becomes the scoring rubric for everything that follows.
- Find the last three things your role delivered. Read the documents, look at the numbers, ask what was hard about them. You are reverse-engineering the standard.
- Set the feedback cadence now, while it is cheap. Ask for a 20-minute check-in every two weeks. Asking in week one is normal; asking in week ten sounds like a problem.
The question people are afraid to ask
"What would success look like?" feels naive. It is the opposite — it forces your manager to commit to criteria in writing-ish form, which protects you later. If the answer is vague ("just be proactive"), push once: "Proactive in what way — flagging risks early, or picking up unassigned work?" You are not being difficult; you are converting a mood into a measurable expectation.
Weeks two to six: build a paper trail, not a highlight reel
Here is the non-obvious insight: your manager is not evaluating your work, they are evaluating their memory of your work. Those are different objects, and the second one is much easier to manage.
A simple weekly habit fixes this. Every Friday, spend ten minutes writing a short log: what you finished, what changed because of it, what is blocked and who can unblock it. Three to five lines. Send it to your manager — not as a brag, but as a status note they can skim.
Why it works: it creates a written record that survives a busy quarter, and it repeatedly answers the question your manager will be asked about you — "how is the new person doing?" A manager with a folder of Friday notes answers instantly. A manager without one guesses.
What belongs in the log:
- Finished, not started. "Sent the supplier comparison, contract signed with option B" beats "working on supplier comparison."
- Effect, if you can name it honestly. "Cut the report prep from two days to half a day by reusing the template." If you cannot measure it, describe what changed in plain words.
- One blocker with a name attached. "Waiting on access to the billing export — asked IT on Tuesday." This is not complaining; it is showing you escalate early.
This is exactly where routine work can be handed off. If your week involves digging through email for a signed document, parsing a contract for a deadline, or drafting the same status update in different words, an AI twin such as the one at asibiont.com can search the mailbox, pull the attachment, extract the amount and date, and draft the note — you still decide what goes in the log and what to send. The judgment call stays with you; the fetching does not.
Weeks seven to twelve: turn results into an argument
By week eight you should have six to eight Friday logs. That is not a diary — it is the raw material for the conversation you are about to have. The mistake here is waiting for the review to be scheduled and then improvising.
Do this instead, three to four weeks before probation ends:
- Ask for the review meeting yourself. "Probation ends on the 14th — can we book 30 minutes this week or next to go through how it has gone?" You control the timing and the framing.
- Prepare a one-page summary. Three columns: what was expected, what happened, what it changed. No adjectives. Numbers where you have them, plain descriptions where you do not.
- Separate the two conversations. The first is "did I pass probation." The second — scope, level, money — is a different talk, ideally two to four weeks later. Mixing them makes the first one tense and the second one rushed.
The trade-off nobody names
There is a real cost to being this systematic. Colleagues who self-promote loudly sometimes get noticed faster in the short term, and a visible paper trail can read as political in a small, informal team. What you gain is defensibility — your case survives a change of manager, a reorganisation, a budget freeze. What you pay is some social friction and about an hour a week. In a company with formal review cycles, that trade is almost always worth it. In a five-person startup where the founder decides everything by feel, a lighter version — the weekly note, no formal dossier — usually works better.
How to ask for a promotion with facts, not promises
When the second conversation happens, structure it in three sentences and then stop talking:
- What changed: "In 90 days I took over the monthly reporting and cut it from two days to half a day."
- What it is worth: "That is roughly a day and a half a month back for the team, and the last two reports went out without corrections."
- What you are asking for: "I would like to move to [level] with [range]. What would need to be true for that to happen this quarter?"
The last question is the important one. It either gets you a concrete answer — a list of conditions you can meet — or it reveals that no conditions exist and the answer is simply no. Both are useful. A "no" with reasons is a plan; a "no" without reasons is a signal about the company.
Expect one of three responses, and know what each means:
| Response | What it usually means | What to do next |
|---|---|---|
| "Yes, at the next cycle" | Budget or timing, not doubt about you | Get the cycle date and the criteria in writing |
| "Not yet — here is what is missing" | A genuine gap, or a vague stall | Ask for the gap in measurable terms and a date to review |
| "We do not have a level above this" | Structural ceiling | Ask about scope, bonus, or a timeline for the role to be created |
If the answer is the third one and it repeats at the next review, that is not a negotiation problem. That is a career decision, and Changing Careers Without Quitting Your Job is often the cheaper path than waiting for a structure to change.
Where this plan breaks — and how to spot it early
The mechanism fails in specific conditions, and it is worth knowing them before you are three months in.
Your manager changes. A new manager inherits you with no memory of your first two months. This is the single most common reason a well-documented probation still ends in "let's wait." Early warning: your manager mentions a reorganisation, or your check-ins quietly stop. Fix: send your one-page summary to the new manager in week one of their tenure, framed as "here is where things stand."
The success criteria move. You were hired for one thing; the team now needs another. Your Friday logs prove you delivered the original brief, but they do not prove you delivered the new one. Early warning: the goals you wrote down in week one are no longer mentioned. Fix: renegotiate explicitly — "the priorities shifted in month two; here is what I did before and after."
Probation quietly expires. In many companies, a probation period that is not formally closed simply rolls into permanent employment by default — but in others, an unclosed review sits in a drawer and surfaces at the worst moment, usually during a budget cut. Check your contract and ask HR directly: "Is the probation formally closed on your side?" This is a five-minute email, and it is the step people skip.
You documented activity, not outcomes. A log full of "attended meetings, reviewed documents" is worse than no log — it shows effort with no result. Early warning: your Friday notes have no verbs like shipped, fixed, cut, closed, decided. Fix: rewrite the last month's entries around what changed.
What depends on your situation
The generic advice — work hard, be visible, ask for feedback — fails differently depending on where you are.
In a large company with formal grades, the promotion is a committee decision made on a calendar, and your manager is an advocate, not a decider. Your job is to give them a case they can paste into a form. Numbers and dates matter more than relationships here.
In a small company, there may be no level to promote into. The realistic ask is scope and money, not a title, and the conversation happens whenever the founder is in a good mood — which is a worse process but a faster one.
In a remote team, visibility does not happen by accident. Nobody sees you stay late; they see what you send. The Friday note is not optional there — it is the only channel your work travels through.
If you are changing field entirely, probation is a different test: you are being judged on speed of learning, not on output. Say so out loud in week one — "judge me on how fast I close the gaps, not on how much I already know" — and then make the gaps visible as they close.
On the tooling side, the parts of this that are pure admin can be automated without touching the parts that require you. An AI twin can set a recurring Friday reminder, pull last week's sent mail and calendar into a draft status note, search the archive for the document you promised someone, and keep the one-page summary updated as entries come in. It cannot decide what your result was worth, cannot sit in the review meeting, and cannot make the ask — that is still a conversation between two people, and it should be.
Start with the one question in week one — "what would make this a clear success in 90 days?" — and the Friday note. Everything else in this plan is just those two habits, repeated and written down. If you want the fetching, searching and drafting off your plate while you focus on the actual work, hand the AI twin your calendar, mailbox and document storage at asibiont.com and let it keep the paper trail warm.