Where notes live
Your client sends video feedback by email.
Here is what that costs, and what to send instead.
A paragraph of notes with no timecodes is not feedback yet. It is a puzzle the editor solves before any cutting starts. This page decodes a realistic email note line by line, shows why the second round is worse than the first, and gives you the short message to send a client so the next set of notes lands on the frame.
Video hosting for filmmakers · 5 GB free · Paid plans from USD 9/month
Updated September 2026
The real cost
An email note hands you a tenth
of the work of receiving it.
Here is a note a client actually sends. It is not a bad note. The person writing it is being helpful, is watching on a laptop between meetings, and has no idea what happens next.
The email, verbatim
“Hey, watched it, looks great overall. The beginning feels a bit slow, maybe tighten it. Also the music is too loud over the interview. The logo at the end should be bigger. Marc says the colour looks a bit green in the second half. Oh, and ignore what I said about the beginning, Sophie likes it as is.”
Reading that takes fifteen seconds. Acting on it does not. Before a single edit happens, someone has to convert six sentences into a list of frames, and every one of those conversions is a guess that can be wrong in a way that costs another round.
Every reading in the left column is plausible from the sentence alone. That is the problem: nothing in an email body narrows it, so the narrowing is done by the person who is supposed to be cutting.
6
Notes in one paragraph
None of them names a frame
0
Timecodes
Every location is reconstructed by hand
3
People with an opinion in it
Two of them are quoted second-hand and cannot be asked a follow-up in the same thread
1
Note that cancels an earlier one
Placed last, after the sign-off, where it is easiest to miss
None of that reconstruction is edit time. It is archaeology, and it does not end when you finish reading. It ends when the follow-up email is answered, which on a job with three stakeholders is the next day. The cost of an emailed note is rarely the note. It is the round trip the note forces.
Round two
Email has no state,
so the second round starts from memory.
The first round survives on goodwill. The second round is where email stops being a workaround and starts being the reason a project slips. A mailbox stores messages. A revision needs something a mailbox has never stored: the current status of each individual note.
Nothing is marked done
You fixed four of the six. The client has no way to see which four, so the reply to your new export repeats two of them and adds a new one, and you read all nine sentences again to work out which are actually open. There is no checkbox in an email body, and there never will be.
Nothing separates round two from round one
Notes on V1 and notes on V2 land in the same thread, in the same font, from the same person. The only marker of which cut a note was written against is the timestamp of the message, and that breaks the moment somebody replies to the older email because it was easier to find.
Reply-all produces contradictions with no referee
Three stakeholders on a thread generate three sets of notes, two of which disagree. The client's producer wants the intro trimmed, the brand lead wants the product held longer, and both are correct in isolation. Nothing in the thread reconciles them, so the editor becomes the referee of somebody else's internal disagreement.
The thread is not addressable by frame
Six months later, the question is what was asked for at 01:14 and who asked. Searching a mailbox finds messages containing words, not notes attached to a moment. The answer becomes a memory, and a memory is not a record.
All four failures share one root. An email note is a sentence with no attributes. Every attribute a revision actually needs, the frame, the version, the author, the status and the round, has to be added afterwards, by hand, by the person with the least spare time on the project.
The replacement
A note is a record with a frame,
not a sentence in a body of text.
The fix is not a nicer inbox. It is giving the note somewhere to live that already holds the attributes an editor needs, so nobody has to reconstruct them. Here is the mechanism, described as what it stores rather than as a feature list.
A timecode in seconds, stored with the note
Not a paragraph mentioning a time. The note is stored against the film at a position, so opening it seeks the player to that moment. That single attribute removes every row of the decode table above: there is no beginning to interpret, because the note sits on the frame the person was looking at when they wrote it.
A drawing on the frame when words will not do
A reviewer can draw directly on the image, and the drawing is saved with the note. A circle around one of three logos ends the entire exchange about which logo. Positional notes are where written language is weakest and where a mark on the picture is worth four sentences.
Replies that thread, and a done flag that closes the note
A question about a note sits under that note instead of at the bottom of a mailbox. When the change is made, the note is marked done. The round is finished when the last one is checked, which is a state anybody can look at, rather than the moment when everyone happens to stop replying.
Rounds, and an approval that can be revoked
Notes group into review rounds, so round two is a set rather than a recollection. Approval is recorded per version, by name, and the film sits in one of three states: reviewing, changes requested, or approved. Posting a new note on an approved cut revokes that approval, which is the honest behaviour: nothing should read approved while an open note sits under it.
The client side of this is deliberately thin. They open a link, play the cut, pause, and type where they paused. There is no account to make and no seat to buy, which is covered properly on its own page. What matters for this page is not how they get in. It is that the note they leave arrives with a frame attached.
Into the timeline
The notes arrive as markers, red for open,
green for done.
A note that lives in a browser and gets retyped into a notepad has solved half the problem. The other half is the trip from the room to the timeline, and that trip is where notes get dropped.
A Premiere Pro panel and a DaVinci Resolve script both exist and are downloadable. They pull the room’s notes over the network and write native timeline markers at the timecode each note carries. Red marks an open note. Green marks one that has been marked done. You work the timeline the way you already do, and the list of what is left is visible on the ruler rather than on a second monitor.
The panel tags its own markers. That means refreshing after a new round updates those markers in place, and never touches the markers you made yourself while cutting. The Premiere panel is a signed extension, and the signature is the studio’s own.
This is the part that turns the change from a client convenience into an editor convenience. Timecoded notes are pleasant for the person writing them. Native markers in the sequence are what actually shortens the round, because the reconstruction step disappears entirely: the note is already at the frame, in the software where the fix happens.
The practical payload
The message to send a client
who keeps emailing notes.
Most editors never make this change because the conversation feels like a complaint about how the client works. It does not have to be. Send this once, at the start of the next round, and say nothing else about it.
Copy this, change the two brackets
“Hi [name], one small change for this round. Instead of replying here, use this link: [link]. Play the cut, pause wherever something needs to change, and type the note right there. It attaches to the exact frame, so nothing gets lost in translation, and I can see which notes are still open and which are handled. There is nothing to install and no account to create. If it is easier for you to send notes another way, do that and I will add them to the page myself, so we both work from the same list.”
Three things are doing work in that paragraph. It gives a reason that benefits the client rather than you, which is that their note lands where they meant it. It removes the objection before it is raised, because the two things clients push back on are installing something and creating an account. And it ends by explicitly permitting them to ignore the whole thing, which is what stops the request from reading as a rule.
Send it with the link, not before it. A message that announces a new process and then asks the client to wait for the cut gets a shrug. A message that is the cut, with one line about where notes go, gets used.
Where this does not work
Some clients will never use it, so
you transcribe, and the record still exists.
A tool does not convert everyone, and a page claiming otherwise is lying to you. There is a category of client, often the most senior person on the account, who will reply to your email from a phone regardless of what you send. Arguing with that is a way to lose a client over a workflow preference.
The workable answer is that you transcribe the email into the room yourself. Open the cut, paste each sentence as a note at the frame you believe it points at, and where you are unsure, put the note at your best guess and add a reply asking which of two moments was meant. It takes about ten minutes for a paragraph like the one at the top of this page. What you get for those ten minutes is the same structure everyone else’s notes have: a frame, a version, a done flag and a round, on your side, permanently.
The side effect is worth naming. Transcribing forces you to resolve every ambiguity immediately rather than at the moment you sit down to cut, and the two or three you cannot resolve become one specific question instead of a vague follow-up. That is a better email than the one you would otherwise send, and it usually arrives the same day rather than the next.
There is also a size below which none of this is worth doing. One note, one person, one short job, and an email is faster than any process. Keep email for that. The threshold is the second stakeholder or the second round, whichever comes first, because both of them need state, and state is the one thing a mailbox has never had.
Questions
Frequently asked
Why is an emailed note worse than the same note in a review room?
The words can be identical. What differs is everything attached to them. A note in the room is stored against the film with a timecode in seconds, so opening it seeks the player to that frame. It carries a done flag, so the round closes when the last note is checked off. It threads, so a follow-up sits under the note it answers instead of at the bottom of a mailbox. An email carries the sentence and nothing else, and the editor rebuilds the rest by hand.
What if my client refuses to use a review tool?
Some will refuse, and no page should pretend otherwise. The workable answer is that you transcribe their email into the room yourself. Paste each sentence as a note at the frame you believe it points at, then reply in the room asking them to confirm the two you were unsure about. It costs about ten minutes and it converts a paragraph into a checklist that exists on your side, with a timecode, a done flag and a round.
How do the notes get into Premiere Pro or DaVinci Resolve?
A Premiere Pro panel and a DaVinci Resolve script both exist and are downloadable. They pull the room's notes over the network and write native timeline markers, red for an open note and green for one marked done. The panel tags its own markers, so refreshing updates them in place and never touches markers you made yourself.
What stops round two from inheriting round one's notes?
Two mechanisms. Every note carries a done flag, so a resolved note stops counting as open. And notes group into review rounds, so round two is a set you can look at rather than a memory of which paragraph arrived when. Posting a new note on a cut that was already approved revokes that approval, which is deliberate: the state should never say approved while an open note sits under it.
Does the client need an account to leave a note?
No. The share link is itself the authentication, and a reviewer never consumes a seat. That is a whole subject of its own and it is covered on the guide about sharing a video with no client account.
Can a client draw on the frame instead of describing it?
Yes. A reviewer can draw on the image and the drawing is saved with the note, which is the fastest way to settle anything positional. A circle around the lower third removes an entire email exchange about which graphic was meant.
Do I lose anything by moving notes out of email?
One real thing: an email thread sits in a mailbox you control, and it is trivially forwardable to a lawyer, a producer or an accountant. Notes in a room live in the room. There is also a size below which email genuinely wins: one note, one person, one small job, and a link is ceremony. The moment there are two stakeholders or a second round, the arithmetic flips.
Send one link and watch where the notes land
Upload a browser-compatible export, share it, and let the next round arrive as timecoded notes with a done flag instead of a paragraph. The review room is on every plan, the free one included.