Useful feedback tells the editor three things: where the problem is, what feels wrong, and what you want the viewer to feel or do instead. We read client notes every day, and the gap between a two round project and a five round project usually comes down to the notes, not the editor. Here is how to write the kind that gets fixed on the first try.
Watch it once before you write anything
Play the full draft start to finish without pausing, the way a viewer would. Your first reaction to the pacing, the hook and the ending is the most valuable thing you can give an editor. You lose it the moment you start scrubbing frame by frame.
Then watch again and write notes. On the second pass you will often find that several things that bugged you were one problem. A slow opening makes everything after it feel slow, so fix the cause and the other notes disappear.
What does a good note look like?
A good note is pinned to a timecode, names the problem and states the goal. "0:42, this pause after the question drags, I want the answer to land faster" gives the editor a target and leaves room to solve it well. They might trim the pause, cut to a reaction or cover it with b roll, the supporting footage laid over the main shot.
A note should tell the editor what is wrong, not which button to press.
Compare that with "make it pop" or "speed this up." Both are honest reactions, but the editor has to guess where and why, and guessing costs a round. Check each note against this list before you send it:
- Where: a timecode or a range, such as 1:10 to 1:25.
- What: the specific element, such as the music, a lower third, a cut, a caption or the color.
- Why: what feels off, in your own words.
- Goal: what you want the viewer to feel or do instead.
- Priority: must fix or nice to have.
Use the review tool your editor sends. Comments in Frame.io or a similar tool sit on the exact frame, so nobody has to translate "the part where she laughs" into a timestamp. A long email of notes with no timecodes is the slowest format there is.
Who should give the notes?
One person. Collect feedback from your team, settle the disagreements internally and send a single set of notes. When the founder wants the intro shorter and the marketing lead wants it longer, the editor cannot pick a winner, and sending both burns a revision.
Conflicting notes from two people are not feedback. They are a meeting you skipped.
Bring every stakeholder in at the first draft, not the third. A compliance reviewer who sees the video for the first time at the end can force changes that undo work you already approved.
How do I describe something I don't have words for?
Point to an example. A video whose pacing you like, or a note saying "music like the track in our spring ad," says more than any adjective. Editors think in references, and "warmer, like the last testimonial" is a note we can act on in minutes.
If you only have a feeling, say so and say where it starts. "Something feels off from 2:00 onward and I can't name it" is useful. A good editor will rewatch that stretch and usually find a jump in audio level, a mismatched shot or a music edit landing on the wrong beat.
What should I leave out of a round?
Match your notes to the stage. On a rough cut, the first assembly of the story, comment on structure: order, what to cut, what is missing. Notes on font size or color grade at that point are wasted, because those shots may not survive the next version.
Flag new ideas as new ideas. If you decide in round two to add a section the brief never mentioned, that is a change in scope, not a fix, and it is fair for it to take more time. Say it plainly so the editor can plan for it.
Say what works, too. "Keep the J cut into the interview at 0:15" (a J cut is where the next shot's audio starts before its picture) tells the editor what to protect while changing other things. It also teaches them your taste for the next video, which compounds when you work with the same editor every time. That is why Crispy Edits pairs each customer with one dedicated editor on every request.
Watch once, write notes with timecodes and goals, send one consolidated round, and name what to keep. Do that and revisions become a quick pass instead of a negotiation.
