Every spring, the same form shows up: describe your growth this year, reflect on your goals, provide evidence of impact. And every spring, teachers either write a diary entry ("This year was challenging because...") or reach for phrases like "leveraged data-driven instruction to maximize student outcomes" that nobody, including the evaluator, actually believes.
The two failure modes
The diary version tells a story but skips the evidence. It's honest, but an evaluator reading twelve of these in an afternoon can't tell your year from anyone else's.
The buzzword version has the right vocabulary but no substance underneath it. It reads like it was written to satisfy a form, because it was.
What actually works is closer to a lawyer's brief: a specific claim, backed by a specific piece of evidence, stated once and moved past.
Building the evidence list first
Before you write a single sentence of prose, build a raw list. Don't worry about phrasing yet.
- What's one thing you changed about your teaching this year, and why?
- What's a piece of student data (not necessarily test scores — could be participation, work samples, behavior referrals) that shows the change had an effect?
- What's a goal you didn't hit, and what did you learn from missing it?
- What's one thing a colleague, parent, or student said that stuck with you?
This list is ugly and disorganized. That's fine — it's the raw material, not the draft.
Where AI actually helps
Once you have the raw list, AI is good at turning bullet points into evaluator-appropriate prose without inventing anything you didn't already tell it. Paste your list along with the actual prompts from your district's form, and ask for a draft that:
1. Opens each section with the specific claim, not a topic sentence about "growth."
2. Attaches the evidence you gave it, not generic phrasing.
3. Keeps the goal-you-missed section honest instead of spinning it into a hidden win.
What it's bad at: inventing evidence you didn't give it. If a draft comes back with a stat or an anecdote you don't recognize, cut it. This document has your name and your evaluator's signature on it — nothing in it should be something you'd struggle to back up if asked in the room.
Our [PD plan generator](/pd-plan) is useful here too, especially if next year's goals section is where you're stuck: it takes your current role and available time and builds the milestone structure your self-evaluation can point back to, so this year's evidence and next year's plan are actually the same document in two tenses instead of two separate exercises.
The section that's hardest to write
Almost every form has some version of "what would you do differently." Teachers either skip it or write something so vague it's meaningless ("I would communicate more"). The honest version is specific and slightly uncomfortable: name the thing that didn't work, name why, and name the one change you're actually making, not five.
An evaluator who reads a real answer here trusts the rest of the document more. It's the paragraph that proves you're not just filling out a form.
What still needs to be true
The self-evaluation is only as good as what actually happened in your classroom this year. AI can help you find the clearest way to say it and make sure you're not underselling work you did, but it can't manufacture a year you didn't have. If the honest draft is thin in places, that's information too — for you, not just for the form.
