The most expensive moment in a revision round is not a difficult note - it is the same note arriving three times through three channels, or two stakeholders giving opposite instructions that you discover only after cutting both directions. Consolidation is the mechanical pass that prevents both: everything lands in one list, conflicts surface before the timeline, and the round ends with a reply everyone signed. It takes twenty minutes and it is the difference between an editor and a note-archaeologist.
Step 1: collect everything into one list
Open one document and pull in every note from every channel: review-tool comments (with their exact timecodes), the email thread, the chat messages from after the call, the call notes themselves. Each entry carries three fields: the timecode or section it refers to, the reviewer's name, and the channel it arrived on. The fields matter later - "which version" and "who said it" are the two questions that decide disputes. If feedback already lives in one review tool with comments pinned to frames, most of this step is export-and-paste; if half of it lives in email, this step is where the leaks stop.
Step 2: merge duplicates
The same note arrives as "the logo part feels off" (email), "logo section - pacing" (chat), and a frame comment at 0:42 saying "tighten this" (review tool). Merge them into one entry, keep the sharpest phrasing, and keep all three citations. Weight comes free with the merge: a note three people made independently outranks a note one person made once, and knowing that changes the conversation when the round runs long.
Step 3: tag every note - fix, preference, or question
Fix: factual or specification failures - the wrong logo, a dead pixel, levels below spec. Preference: taste calls - pacing feel, music choice, grade mood. Question: anything that needs an answer before work. Then apply priority on top: MUST, SHOULD, COULD (the same tags the client-facing etiquette guide teaches - see client feedback etiquette). The discipline is in refusing to let preferences masquerade as fixes: naming the distinction is what keeps rounds inside contract, and the round-count mechanics are in how many revision rounds are normal.
Step 4: reconcile conflicts before the timeline
Two stakeholders want opposite things - "cut the intro shorter" against "the intro is the hook, keep it." The rule is structural: conflicts are reconciled by a decision owner before any editing happens, not arbitrated by the editor mid-cut. Send one short message: "Notes 4 and 7 conflict - which direction? One decision, please, by Thursday." The editor's job is to surface the conflict, not to guess who wins. If your project has no named decision owner, that is the actual bug - fixing it is a contract clause, and the client-side habit that prevents the conflicts in the first place is one consolidated set of notes per round (the templates for that message are in video feedback comment templates).
Step 5: sort into a fix list grouped by timeline position
Order the surviving notes by where they land on the timeline - 0:00-1:00, 1:00-2:30, and so on - and mark what this round will address versus what is explicitly deferred. The grouped list is your cutting plan: you make one pass through the timeline instead of eight interrupted ones. Deferred notes go back to the client in writing so "you ignored note 9" cannot happen - the round-boundary language for that reply is in video review email templates.
Step 6: reply with the round summary
Close the loop in writing before cutting: "Round 2, 11 notes consolidated to 9 (two duplicates). Addressing 7, deferring 2 (music licensing timing, intro length - flagged as conflicting, awaiting your call). New version Thursday." The summary converts your consolidation into shared reality, and it is also the record that proves the round was scoped - which is the entire mechanism that keeps round five from looking like round one (the checklist that keeps rounds tight is in the video review checklist).
Tool reality: a review tool with frame-pinned comments, version stacks, and resolved threads does most of steps 1 and 2 for you - comments are already at exact frames with names attached, so consolidation shrinks to dedupe, tag, and sort. The failure mode is not the tool; it is the notes that bypassed it. The receive-side habits that keep feedback on-frame are in how to receive client feedback on video, and the vocabulary of exact-frame addressing is in what frame-accurate review means.
FAQ
Who should consolidate client notes - the editor or the client?
The client side consolidates their own stakeholders' notes into one set; the editor (or producer) consolidates across channels into the fix list. A single editor reconciling three stakeholders' raw opinions mid-cut is the failure shape - the "single consolidated set per round" rule is client-side etiquette, and enforcing it is a producer conversation, not a cutting-room one.
How do I handle two stakeholders who want opposite changes?
Surface it before cutting: one message naming both notes and asking for one decision from the named owner, with a deadline. Never split the difference in the timeline - both parties end up unhappy and the round doubles. If no decision owner exists, creating one is the fix; the structure lives in the approval workflow guide.
What do I do with notes that arrive outside the review tool?
Capture them into the list with their channel and author, then answer them where the record lives: "noted from your Tuesday email - logged as note 6, will address in this round." The point is not scolding the channel; it is making sure the fix list is complete, because an un-captured note is a future "you ignored this" - the send-side ask that prevents the leak is in client feedback etiquette.
When does a note stop being a fix and become scope creep?
When it changes the agreed deliverable rather than correcting it - a new section, a restructure of the approved structure, a second deliverable born from the notes of the first. The tag in step 3 exists to catch exactly this: a "fix" that no specification failure backs up is scope, and scope is a conversation with a price, not a silent edit. The round-count contract that makes the boundary real is in how many revision rounds are normal.
How long should consolidation take?
Twenty minutes for a typical client round (up to an hour for a multi-stakeholder corporate review with 30+ notes). If consolidation routinely takes longer, the upstream problem is channel chaos, not your process - more tools consolidated harder, or one review link with named lanes (the reviewer-lane pattern is in video review best practices).