Observation Copilot can generate structured, framework-aligned feedback from virtually any style of notes - but like any tool, the output quality depends on the input. Principals who write more detailed, evidence-based notes consistently get more specific, more useful feedback drafts.
These tips will help you write observation notes that get the most out of AI-powered feedback generation.
Tip 1: Capture What You See and Hear, Not What You Think
The most common note-taking pitfall is jumping straight to evaluation. Instead of writing "good questioning techniques," describe what actually happened:
- Instead of: "Teacher used effective questioning"
- Write: "Teacher asked 'What evidence from the text supports your claim?' - 4 students raised hands, teacher called on a student who hadn't participated yet, then asked a follow-up: 'Can anyone build on that?'"
The more concrete evidence you capture, the more specific the AI-generated feedback becomes. Observation Copilot can turn raw evidence into framework-aligned analysis, but it can't invent details you didn't record.
Tip 2: Use Timestamps
Adding rough timestamps to your notes helps in two ways. First, it shows the pacing and flow of the lesson. Second, it gives the AI context about transitions and time allocation:
- "9:05 - Warm-up on board, students working independently, teacher circulating and checking in with 3 students individually"
- "9:15 - Transition to whole-group instruction, took about 2 minutes, students had materials ready"
- "9:25 - Small group work begins, teacher pulls a guided reading group of 4 students"
You don't need to be precise - even rough timestamps every 5-10 minutes create a useful picture of the lesson.
Tip 3: Note Student Behavior, Not Just Teacher Behavior
Many principals focus exclusively on what the teacher is doing. But student behavior is some of the strongest evidence for evaluation frameworks:
- "During independent practice, ~80% of students were on task. Two students in back corner were off-task, teacher redirected with proximity."
- "During discussion, 6 out of 22 students participated verbally. Others appeared to be listening but weren't engaged in the conversation."
Student engagement, responses, and behaviors provide the evidence that frameworks like the Danielson Framework for Teaching are designed to evaluate.
Tip 4: Don't Worry About Organization
One of the biggest advantages of using Observation Copilot is that you don't need to organize your notes by framework domain while observing. Just write what you see, in order. The AI handles the sorting.
That frees you to actually watch the classroom instead of splitting your attention between observing and figuring out which rubric domain each note belongs to.
Let the tool do the organizing. You focus on seeing what's happening.
Tip 5: Include Context When It Matters
A brief note about context at the top of your observation can improve the quality of the generated feedback:
- "3rd grade ELA, 22 students, co-taught with SPED teacher, lesson on main idea and supporting details"
- "This is the teacher's 2nd year, working on improving differentiation based on last observation cycle"
Context helps the AI generate next steps that are relevant to the teacher's situation rather than generic recommendations.
The Bottom Line
You don't need to change how you take notes dramatically. The biggest wins come from capturing specific evidence (what was said and done) rather than judgments, and from not worrying about organizing notes into framework domains while you observe. Observation Copilot handles the structure - you handle the seeing.
Try it with your next observation at app.observationcopilot.com. Paste in your notes and see the difference detailed evidence makes.
Paste your notes and see the difference detailed evidence makes.