You can still write a strong self-review with no notes: reconstruct it from the trail you already left behind. You did not keep a log, but your calendar, your merged pull requests, your closed tickets, your sent email, and your Slack history all recorded the year for you. Walk the review period from the start, pull every real thing into a dated list, pick your five strongest, and write each as a STAR bullet with a number attached. You do not need to remember the year; you need to read it back from the tools that already have it. Then fix the actual problem so the next review is not the same scramble: start capturing one line a day from today.
One thing to be clear about up front: nothing here recovers notes you never took, because nothing can. What it does is find the evidence that already exists in the systems you use every day and turn it into review-ready lines. That is a better position than it feels like right now, because the trail is usually more complete than your memory ever was.
Where your year is already written down
You have six sources, and between them they hold almost everything you did. Each answers a different question, so cross-referencing two or three fills the gaps fast. Work the table top to bottom: open the source, search it for what it is good at, and turn what you find into one dated line.
| Source | What to search for | What it becomes |
|---|---|---|
| Calendar | Launches, milestones, recurring meetings you ran, offsites, interviews you did. | What you were in: “Ran the weekly roadmap review all year”; “Led the March launch readiness meetings.” |
| Git & pull requests | Merged PRs you authored, PRs you reviewed, the big branches you owned. | What you shipped: “Merged the onboarding-flow rewrite in March”; “Reviewed 60-plus PRs this half.” |
| Tickets & project tracker | Issues closed or delivered, epics you drove, incidents you were on. | Delivered work with dates: “Closed the billing-migration epic, zero rollbacks.” |
| Slack & chat | Search your own name and the word “thanks”; decisions that landed in a thread. | What you drove and the praise you got: “Priya thanked me for the API docs in #launch (April).” |
| Sent email | Proposals, status updates, decks you sent, threads where you moved something forward. | Influence and communication: “Sent the reliability-sprint proposal that shipped in Q3.” |
| 1:1 & goals doc | What you committed to at the start of the period; check each goal against what actually landed. | Goals versus outcomes: “Committed to owning the data layer; shipped two database-heavy projects.” |
The calendar and the git history are usually the richest two, so start there. If a source is empty for you (not everyone lives in Slack or git), skip it and lean harder on the ones that hold your work. The goal is not to fill every row. It is to gather enough dated, real moments that your five best pick themselves.
Do this in the next hour
You don’t need to remember the year. You need to read it back from the tools that already recorded it. Work in this order so you’re not staring at a blank form.
- Open your six sources in tabs. Calendar, git and PR history, your ticket tracker, Slack (search your own name and “thanks”), sent email, and any 1:1 or goals doc. Set the date range to the review period and start at the beginning, not last week. Recency is the trap.
- Dump everything raw. Paste one dated line per real thing into a scratch list. Don’t judge it yet. “Mar 9: merged the onboarding-flow PR.” “Apr: Priya thanked me for the API docs in #launch.” Volume first.
- Pick your five strongest. No more. Impact, not effort.
- Write each as STAR (Situation, Task, Action, Result) with the number or outcome attached.
- Add one honest growth area with the plan you’re already following. Done.
Turn a reconstructed fact into a review line
A raw line from your calendar or PR list is the evidence, not the review. The last step is phrasing: take the dated fact and attach the impact, so it reads as a result instead of a task. Here is the same move the self-evaluation examples page makes at length. Watch vague turn specific.
- Vague: “Helped ship the billing migration.” Specific: “Led the billing migration end to end across two services. Zero downtime, zero rollbacks, and it unblocked three teams waiting on it.”
- Vague: “Improved our reporting.” Specific: “Rebuilt the weekly metrics pipeline so the report runs itself. Saved the team about four hours a week and killed a recurring Monday fire drill.”
- Vague: “Did a lot of code reviews.” Specific: “Reviewed 60-plus PRs this half and unblocked two stuck projects by pairing with the owners instead of leaving comments and waiting.”
- Vague: “Communicated well.” Specific: “Rewrote the onboarding docs after new hires kept asking the same five questions. Ramp time dropped from about three weeks to ten days.”
The dates and numbers in these examples are illustrative patterns, not claims about anyone. Swap in your own reconstructed facts, the ones you can stand behind, and the line does the work.
See how a self-review reads back from evidence instead of memory.
The copy-paste rescue template
Drop your reconstructed facts straight into this. Each bucket is pre-labelled with the source that feeds it, so you know exactly where to go dig. Fill the buckets, pick your top five, STAR each one, add a growth area, and you have a draft.
SELF-REVIEW RESCUE | [Name] | [Period] Reconstructed from sources I already have. No notes required. FROM CALENDAR (meetings, launches, milestones) - [date]: FROM GIT / PRs (what I shipped or reviewed) - [date]: FROM TICKETS (closed / delivered work) - [date]: FROM SLACK & EMAIL (praise, decisions, "thanks") - [date]: FROM 1:1 / GOALS DOC (what I committed to -> what landed) - [date]: TOP 5 (STAR: Situation, Task, Action, Result + a number) 1. 2. 3. 4. 5. GROWTH AREA + PLAN -
So this never happens again: start capturing today
Reconstruction works, but it is a rescue, not a plan. The reason you are here is that the wins were never written down while they were fresh, and memory is a terrible filing system. The fix is not more discipline at review time; it is a five-second habit the rest of the year. Two rhythms do it:
- In the moment. Right after anything goes well, whether it is a launch, a piece of praise, or a problem solved, drop one dated line before you move on. The details are still in your head, which is exactly when the best entries get written.
- End of day or week. Once a day or once a week, take two minutes to fill the gaps: what shipped, what you decided, what you learned. The weekly pass catches the quiet work that never felt like a “win” in the moment but reads like one at review time.
That is the whole habit, and it is what turns the next review from archaeology into reading. The work-diary guide goes deep on making it stick; the short version is: capture one line, tag who or what it was about, keep it in one place you control.
This is exactly what Notivo is built to make effortless. Write a quick line about what you did and tag it with @ and a person or # and a topic, and it files itself into a dated, private timeline. Tag your own projects and your wins collect in one searchable thread; tag the people and topics involved and you can pull “everything I did on #billing” in one search. Capture the line however you are working: on the web, over WhatsApp, or from the ChatGPT and Claude connectors, so the win gets saved in the five seconds you actually have. When review season comes, an AI assistant can recap your own notes only to speed the read. It reads what you wrote, is consent-gated on paid plans, and never invents wins you did not log. It does not write your self-review. You do, and it is stronger because the evidence is already organized in front of you.
For the lines you would not want anyone reading over your shoulder, whether it is a frank note about a manager, a salary figure, or something genuinely sensitive, Notivo has a Black Box vault you open with @bb or #bb. The Black Box is PIN-locked and end-to-end encrypted, so only you can read what is inside.
On storage for everything else, the honest answer: your notes are private by default and scoped to your own account, and they are stored on secured infrastructure. That is platform-level security, the protections that come with reputable cloud storage. It is not a claim that your regular note content is encrypted end-to-end, and we are not going to pretend otherwise. Only the Black Box is end-to-end encrypted. So use the same judgment you would use with any cloud notes app: this is the right home for your running record of what you did, and the Black Box is there for the rest.
Capture a win in five seconds from web, WhatsApp, or the ChatGPT and Claude connectors.
Log the whole year as it happens, so review season is reading, not archaeology.