Skip to content

Guide

A-roll is the footage the video is made of. Everything else is laid over it

A-roll is the primary footage of a video: the shot that carries the story and the audio the whole edit runs on. In a vertical ad it is the person speaking to camera. B-roll is everything cut in over it. The names come from the two physical rolls of film editors once ran in parallel.

Continuous take the whole edit is built on
1
Of the runtime the A-roll keeps by pace
48-60%
Source ceiling for a take here
3 min

The audio is the spine, and the A-roll owns it

Strip a modern vertical ad to its skeleton and what remains is one continuous piece of speech. That speech, and the shot it was recorded with, is the A-roll.

Everything else in the edit is subordinate to it. Cutaways replace the picture while the A-roll audio keeps running. Captions transcribe it. Music sits under it.

This is why the terms are not symmetrical. B-roll can be swapped, trimmed or deleted and the video survives. Cut the A-roll and there is no video.

The physical origin explains the names. Film editors ran two rolls of negative in parallel to make transitions possible. The A-roll carried the scene. The B-roll carried the second image. The rolls are gone and the hierarchy stayed.

Good A-roll is an audio decision wearing a video costume

Because the A-roll carries the sound, its quality is mostly heard rather than seen. A sharp image with muddy audio is bad A-roll. A plain image with clean, close speech is good A-roll.

The practical rules follow from that. Record close to the phone. Kill anything with a fan. Prefer rooms with soft surfaces. Speak the whole message in one continuous pass rather than fragments.

One continuous pass matters for the edit, not just the sound. Trims, cutaways and captions all assume a single stream of speech to cut against. Fragments recorded separately never quite join.

The camera is the last concern. A phone at eye height in a real room is the entire equipment list this format needs.

What hangs off the A-roll in a finished ad

One recording, four dependents. Every layer reads its timing from the speech underneath.

  1. The take

    Speech, one pass

  2. The cut

    Trims follow the sentences

  3. The cutaways

    Land on specific words

  4. Captions and music

    Timed to the speech

This is the whole editor

Highlight a phrase and a clip lands on those exact words. No timeline, no keyframes, no layers.

CLIPS · 5I have thisexact conversationeverysingle week. Somebody sits down and says,oh yeah, I takecinnamonevery day.And honestly, doc, I have no idea if it works.So let me tell you what isin that capsule.a clip lands on these wordscut from the editTAKING CINNAMONEVERY DAY?is it doing anythingHeadlineMusicCaptionsTHIS VIDEOLength25.0sClips5Words removed18Export video

How much A-roll must stay on screen has a real answer

A talking ad works because a person is visibly saying something to the viewer. Cover too much of the A-roll with footage and that person disappears, and with them the reason the format converts.

A workable ceiling is cutaways covering under half the runtime, with the face present at the hook, the proof and the ask.

Cutroom enforces this by pace: b-roll may cover 40 percent of the runtime at chill, 45 at normal and 52 at fast, which leaves the A-roll the majority in every mode.

The faster the cutting, the more coverage the ad tolerates, because each clip is shorter and the face keeps returning. The ceiling moves with rhythm, not taste.

The share of runtime the A-roll keeps

The inverse of the coverage caps by pace. In every mode the person stays the majority of the ad.

  • Chill pace60%
  • Normal pace55%
  • Fast pace48%

Record the A-roll like the asset it is

One take produces many ads, because the argument and the opening come apart cleanly. Record the body once and several different openings in the same sitting, and each opening is a new creative.

Keep takes under the ceiling of the tool that will cut them. Here that is three minutes of source, which is roughly 450 spoken words and far more than a thirty second ad needs.

Read the transcript back before exporting anything. The A-roll carries the brand name, the price and the claim, and a transcription error in any of them ships in the captions.

Treat rejected takes as material rather than failures. A strong sentence in a weak take can open the next ad.

Questions people ask

Is A-roll always a person talking?
No. In documentary it is the interview, in a tutorial the demonstration, in narrated video the voice track plus its primary footage. In vertical ads it is nearly always a person to camera, which is the case this page assumes.
Can a video be all A-roll with no b-roll?
Yes, and strong ones often are. A person with a sharp argument holds a feed on jump cuts and captions alone. B-roll earns its place per clip. It is not an obligation.
What ratio of A-roll to b-roll should I shoot?
Shoot the A-roll to completion first, since without it nothing else matters. B-roll needs are set by the script afterwards. Usually a handful of phrases genuinely earn a picture in a thirty second ad.
Does A-roll quality matter more than b-roll quality?
Yes, and it is not close. Viewers forgive rough cutaway footage and do not forgive unintelligible speech. Spend effort on audio and delivery before spending anything on coverage.

The A-roll is the ad. Everything else decorates it. Cutroom takes that one take and hangs the cut, the cutaways and the captions off it, which is the order the work was always in.

3 videos free, no card3 finished videos free in your first 7 days, no card. They carry a Cutroom mark; Lite at $19.99/month removes it