Split the general appraisal into fact plus gaps
“Completed all assigned tasks” becomes the checkable statement of delivering to schedule, with project names, count, and dates listed as still needed.
Turn the notes you already have into a three-part comment: facts, impact, next steps. Facts and opinions are written separately, every subjective judgement is labelled as one, and we do not write anything about promotion or pay.
We extract checkable events and figures, label subjective judgement, set out impact and next steps, and list what you still need to confirm.
SAMPLE OUTPUT
This is the complete output for the source text in the demo above: a comment-quality score, itemised revisions, what was dropped and why, items to confirm, and the drafted comment. It is a demonstration, not a customer case.
“Completed all assigned tasks” becomes the checkable statement of delivering to schedule, with project names, count, and dates listed as still needed.
“Unanimous praise from leaders and colleagues” has no recorded source and cannot be checked, so the whole sentence is removed with the reason stated.
The view that he invests in cross-team collaboration is kept, but is explicitly written as the author's judgement rather than fact, and separated from the facts section.
# Review comment (draft) ## Facts - Completed the assigned development work within this cycle and delivered to the agreed schedule. - Took part in team collaboration and requirement reviews. TODO: add specific project names, delivery count, and dates TODO: add the metric and measurement basis behind “results” ## Impact TODO: the material records no observable change resulting from the delivery; add it before writing this section ## Next steps TODO: to be added, based on specific facts from this cycle ## Opinion (the author's judgement, not a statement of fact) - In my view he invests noticeably in cross-team collaboration, which was observable across several requirement reviews. --- **Not used from the source**: “outstanding results” and “unanimous praise from leaders and colleagues” — the material contains no metric data and records no source for any third-party opinion. TODO: if these must be used, add the metric basis and the source of the evaluation.
WHAT DRAFTING DOES
“Shipped three releases” is a fact; “performed well” is a judgement. Facts come first so the judgement has something to stand on.
After what was done: whose path got shorter, which problems got rarer. If there is no impact, say why the point is worth making.
“I think he is ready to mentor juniors” must read as a view, not as “everyone thinks so”. One mislabelled sentence costs the whole paragraph its credibility.
WHY IT IS WORTH DOING
“Performed well and cooperated actively” tells the reader nothing. The person receiving it cannot tell what to do next.
What lands in a comment gets quoted in later conversations, pay reviews, and disputes. Invent once and everything downstream is wrong.
“Promotion likely next year” reads as encouragement and lands as a commitment — and it is not the commenter's decision to make.
Note: performance management practice generally holds that comments should rest on observable behaviour and results, with subjective assessment recorded separately from fact. Companies and jurisdictions differ in how employee records must be kept and worded. This site organises text only; it is not HR, legal, or employment advice.
DRAFTING DEMO
Xiao Wang performed well this year, completed all assigned tasks, cooperated actively with the team, delivered outstanding results, and received unanimous praise from leaders and colleagues. We hope he keeps it up next year.
# Review comment (draft) ## Facts - Completed the assigned development work within this cycle and delivered to the agreed schedule. - Took part in team collaboration and requirement reviews. TODO: add specific project names, delivery count, and dates TODO: add the metric and measurement basis behind “results” ## Impact TODO: the material records no observable change resulting from the delivery; add it before writing this section ## Next steps TODO: to be added, based on specific facts from this cycle ## Opinion (the author's judgement, not a statement of fact) - In my view he invests noticeably in cross-team collaboration, which was observable across several requirement reviews. --- **Not used from the source**: “outstanding results” and “unanimous praise from leaders and colleagues” — the material contains no metric data and records no source for any third-party opinion. TODO: if these must be used, add the metric basis and the source of the evaluation.
Demo sample: removes conclusions with no data behind them and praise that cannot be traced, keeps the checkable facts, and labels judgement separately. Paste your material to get your version (sign-in required; one credit per run).
WHY YOU CAN TRUST IT
No project name, delivery count, or metric basis in the material means a placeholder — never a plausible-looking figure.
“Everyone says he is great” cannot be written. Either supply the source or leave it out.
We do not write “promotion likely” or “pay rise in view”. Those decisions are not the commenter's, and on paper they are a liability.
Submitted text is used for this single AI request and is not intentionally persisted. See the Privacy Policy.
FAQ
No. Missing project names, counts, and metrics become placeholders. Comments get quoted in later conversations, and the person reviewed pays for anything invented.
Yes, but it is labelled as an opinion and paired with the observable fact behind it wherever possible. Unsupported judgement gets its own section rather than being mixed into the facts.
No. Promotion, rating, compensation, bonus, and continued employment are not decided by a comment, so we neither state nor imply them.
One credit per run, usually done in seconds, returning the comment, the revision notes, and items to confirm.
No. Credits never expire — whatever you do not use carries over to the next cycle instead of resetting.
Not yet. Each run processes one document and consumes one credit; there is no bulk upload or API today. Email us with your use case and we will assess it.
Not intentionally. Text is used for this single AI request only. See the Privacy Policy.
Quarterly or annual reviews, peer feedback, post-project notes, probation comments — any assessment written for the person themselves and likely to be quoted later.
SIMPLE PRICING
For an occasional publishing cadence, with a focused monthly credit allocation.
For frequent publishers who iterate on content continuously.
Our largest monthly credit allocation, for the heaviest continuous usage.
Credits never expire: unused credits carry over instead of resetting. Credits apply to this site only and are not shared with other sites.
Prices shown are USD baselines; final currency, taxes, and discounts are determined by Paddle for the checkout location.