Self-reviews

Self-Evaluation Examples: How to Stop Writing From Memory

A strong self-evaluation answers three things for each competency: what you did, the impact it had, and what you are improving next. Below are 30+ copyable examples grouped by competency, then the habit that means you never write a self-review from a blank memory again.

Tonight's review is the last one you write from memory. Paste your first win now.

Below are 30+ copyable examples grouped by competency: results, communication, collaboration, and growth, each in before/after form so you can watch vague turn specific. But the reason most self-reviews feel thin is not writing skill. It is recall. By review season you have forgotten what you shipped in March. The examples here will help you phrase it. The second half shows the habit that means you log your wins as they happen, so the writing-from-memory part never has to hurt again.

Self-evaluation examples by competency

Steal these directly, or use them as a pattern. Each pair shows the weak version most people write from memory, then the specific version that names a project, a number, or a real outcome. Swap in your own facts. The goal is not to sound impressive. It is to be accurate enough that a reader can picture the work.

Results and impact

  • Vague: "Worked on performance improvements." Specific: "Owned the checkout latency project this half. Cut p95 from 1.4s to 380ms, which reduced cart abandonment 22% in the A/B test."
  • Vague: "Handled a lot of support tickets." Specific: "Closed 14% more support tickets than target while keeping CSAT at 4.7 out of 5."
  • Vague: "Helped ship the billing migration." Specific: "Led the billing migration from start to finish across two services. Zero downtime, zero rollbacks, and it unblocked three teams that were 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 removed a recurring Monday-morning fire drill."
  • Vague: "Hit my sales numbers." Specific: "Closed 118% of my annual quota and brought in two of the three largest new accounts of the year."
  • Vague: "Did a lot of bug fixing." Specific: "Drove the reliability push that cut Sev-2 incidents from nine in Q1 to two in Q4 by fixing the three flaky systems behind most of them."

Communication

  • Vague: "Communicated well with the team." Specific: "Rewrote our onboarding docs after noticing new hires asked the same five questions. Ramp time dropped from about three weeks to about ten days."
  • Vague: "Kept stakeholders informed." Specific: "Started posting weekly project summaries in the shared channel. Two teams told me it cut their status pings to me to almost nothing."
  • Vague: "Presented at meetings." Specific: "Ran the quarterly roadmap review for the whole department, 40-plus people, and turned the deck into the format leadership now reuses each quarter."
  • Vague: "Wrote good documentation." Specific: "Wrote the incident runbook after our payments outage. It has been used in three incidents since and cut average time-to-resolution roughly in half."
  • Vague: "Gave clear updates." Specific: "Replaced our wall-of-text status emails with a three-line format: shipped, in flight, blocked. Open rates went up and follow-up questions went down."

Collaboration

  • Vague: "Worked closely with design." Specific: "Paired with design weekly on the redesign instead of waiting for handoff. We caught issues early and shipped two sprints ahead of plan."
  • Vague: "Helped onboard a new teammate." Specific: "Mentored a new teammate through their first on-call rotation. They were running it solo by week three."
  • Vague: "Was a team player." Specific: "Volunteered to cover the on-call gap when two people were out, then wrote up the playbook so the next gap is not a scramble."
  • Vague: "Worked across teams." Specific: "Coordinated the launch with marketing and support so docs, release notes, and the help center all went live the same day. No scrambling at the finish line."
  • Vague: "Supported other people's projects." Specific: "Reviewed 60-plus pull requests this half and unblocked two stuck projects by pairing with the owners for an afternoon each instead of leaving comments and waiting."

Ownership and reliability

  • Vague: "Was dependable." Specific: "Owned the on-call schedule for the team all year. No missed handoffs, and I rebuilt the escalation doc so a page actually reaches the right person."
  • Vague: "Took initiative." Specific: "Noticed our error budget was quietly burning and proposed the reliability sprint before anyone asked. We came out of it under budget for the first time in three quarters."
  • Vague: "Delivered on time." Specific: "Shipped six of the seven projects I committed to this period. The one slip I flagged early and re-planned, so nothing surprised anyone downstream."
  • Vague: "Handled problems well." Specific: "Caught the data-leak bug in staging before it reached production, which saved a customer-facing escalation, and added the test that now blocks that class of bug."

Leadership and influence

  • Vague: "Showed leadership." Specific: "Led the cross-team launch with no formal authority. Ran daily syncs and a shared tracker, and we shipped on date with zero P1s."
  • Vague: "Set a good example." Specific: "Started the Friday demo habit on the team. Six months in, everyone presents their own work and morale around shipping is visibly higher."
  • Vague: "Influenced the roadmap." Specific: "Made the data case for deprecating the legacy importer. It shipped this quarter and removed about a third of our support load."
  • Vague: "Mentored people." Specific: "Ran a weekly office hour for two junior engineers. Both shipped their first independent feature this half and one is now mentoring the next hire."

Growth and development (honest, not self-flagellating)

  • Weak: "I need to manage my time better." Honest with a plan: "My estimates ran long in Q3. I started doing scope reviews before committing to dates, and my last three projects landed on time."
  • Weak: "I should speak up more." Honest with a plan: "I held back in larger meetings early in the year. I have been pre-writing my main point beforehand, and I have raised a concern that changed a decision twice this quarter."
  • Weak: "I want to get better at delegation." Honest with a plan: "I was the bottleneck on reviews. I documented the review checklist and handed two recurring areas to teammates, which cut my review queue in half without quality slipping."
  • Weak: "I could improve my technical depth." Honest with a plan: "I leaned on others for the data layer. I spent this half owning two database-heavy projects and can now make those calls without escalating."
  • Weak: "I should document more." Honest with a plan: "Knowledge lived in my head, which slowed the team when I was out. I wrote up the three systems only I knew, and the last on-call week ran without paging me once."

The self-evaluation template

Once you have your examples, drop them into this structure. For each competency, write one line on what you did, one on the impact, and one on what is next. Keep it to two or three examples per section. Copy it and fill it in.

Copy the template
SELF-EVALUATION · [Your name] · [Period]
For each: WHAT I did · IMPACT · WHAT'S NEXT

Results:
Technical/Craft:
Communication:
Collaboration:
Growth area + plan:
Goals next period:

Common self-evaluation mistakes

Most thin self-reviews share the same handful of habits. Scan this list before you submit.

  • Vague adjectives. "Hardworking," "dependable," and "great communicator" describe a personality, not a result. Replace each one with a thing that happened.
  • No numbers. A number anchors a claim. "Faster" is an opinion; "1.4s to 380ms" is evidence.
  • Listing tasks, not impact. "Attended planning, wrote tickets, did code review" is a job description. What changed because you did them?
  • Hiding growth areas. A self-review with no honest weakness reads as either unaware or defensive. One growth area plus a plan reads as confidence.
  • Writing from memory. The single biggest one. If you are reconstructing your year the night before it is due, the result will be a blurry highlight reel weighted toward the last few weeks.

Why your self-review feels like a memory test

"Wait, what did I actually do this year?" That blank-mind moment at the start of self-evaluation season is almost universal, and it is not a sign you had a quiet year. It is a sign that your brain is built to solve today's problem, not to keep a running archive of your wins. The work that felt important on a Tuesday in March is gone by review season, buried under everything that happened since.

So most people do memory archaeology: digging back through old tickets, the calendar, and sent messages, trying to reconstruct a year from fragments the week it is due. It is slow, it is stressful, and it systematically undercounts you, because the quiet, steady work that does not leave a dramatic trace is exactly the work memory throws away first. The examples above can make any given win sound sharp. They cannot recover a win you have already forgotten.

Log your wins as they happen, so the self-review writes itself

The fix is not a better template or a more disciplined memory. It is a private log you add to as you go. Drop a one-line note the moment you ship something, tag it by project, and at review time you search the tag and the self-review assembles itself from dated facts you no longer have to recall.

That is what Notivo is for. Write a quick private note when a win lands and tag it #checkout or #onboarding. It files itself into that topic's timeline. Mention a teammate with @Priya and it also lands on their timeline, so kudos and shared wins are easy to find later. When self-review season arrives, you open the tag and read your year back in order, dated and specific, then paste it into the template above. Prep stops being remembering and becomes reading.

Today you can do all of this on the web: create and edit notes, parse @people and #tags into per-person and per-topic timelines, search across everything, and export it whenever you want. Your notes are private by default and account-scoped, stored on secured infrastructure. That last part is platform-level security, the kind that comes with reputable cloud infrastructure, plus account-level access controls that keep your notes visible only to you. We keep the security story honest rather than dressed up.

Related reading

Keep building the habit and the case it supports:

Frequently asked questions

How do I write a self-evaluation?

For each competency, cover three things: what you did, the impact it had, and what you are improving next. Back every point with a dated, specific example instead of a general claim. Two or three strong examples per competency is plenty, and they read far better when each one names a project, a number, or a named outcome.

What if I can't remember my year?

Reconstruct it from the work you already left behind: merged pull requests, shipped projects, closed tickets, sent documents, and your calendar. Walk the year quarter by quarter and write down what you find. Then start logging going forward so next review season you read your wins instead of mining for them.

How long should a self-evaluation be?

Shorter and specific beats long and vague. Two to three strong, concrete examples per competency are more convincing than a long list of generic statements. Lead with impact, keep each example to a sentence or two, and cut anything you cannot tie to a result.

Should I mention weaknesses?

Yes. Naming one honest growth area plus the plan you are following shows self-awareness, and it reads as confidence rather than a confession. Frame it as a behavior you are changing with evidence of progress, not as a character flaw.

How do I make examples specific?

Attach a number, a date, or a named outcome to each one. "Improved onboarding" is forgettable; "rewrote the onboarding docs and cut ramp time from about three weeks to ten days" is specific and defensible. The detail almost always comes from notes you kept during the year, not from memory at the blank page.

Your year, written down as it happens

Stop writing your self-review from memory

Log one win the day it lands, tag it by project, and let review season be copy-paste instead of a memory test. Private by default. Free to start.