Skip to content
Klyp
FramingInterviews

Vertical framing for interviews: follow, split, or fit?

Turning a wide two-person interview into a 9:16 clip throws away most of the frame. Here is how to decide what to keep — and when each layout works best.

The Klyp team3 min read

Footage: ITU · AI for Good · CC BY. Rendered by Klyp.
On this page
  1. The math of a vertical crop
  2. Option 1: Follow the speaker
  3. Option 2: Split screen
  4. Option 3: Fit with a blurred background
  5. Option 4: Center crop
  6. Choosing quickly
  7. Don’t forget the other shapes
  8. Captions and hooks follow the framing

Most interviews and podcasts are filmed wide: two people, side by side or facing each other, in a 16:9 frame. Short-form platforms want 9:16. Getting from one to the other is mostly a question of what you are willing to lose.

The math of a vertical crop

Take a 1920 × 1080 frame. A 9:16 crop at full height is 1080 × 9 ÷ 16 ≈ 608 pixels wide — about a third of the original width. If your two speakers sit at opposite sides of the frame, there is no single crop that shows both of them. Something has to give.

A 16:9 source: the camera cuts between interviewer and guest. Footage: ITU · AI for Good · CC BY. Rendered by Klyp.

Option 1: Follow the speaker

Crop to whoever is talking, and cut to the other person when they answer. Done well, it looks like a deliberately edited two-camera shoot.

  • Works best for: back-and-forth conversations, podcasts, interviews where both people matter.
  • Watch out for: fast cross-talk. If speakers change every second, constant cuts become tiring. Hold on one person through short interjections like “right” or “yeah”.
  • Detail that matters: keep the crop steady while someone talks. Small head movements should not make the frame wobble; only re-centre when the person really moves.
Follow speaker: the crop cuts to whoever is talking. Footage: ITU · AI for Good · CC BY. Rendered by Klyp.

Option 2: Split screen

Stack the two people: one in the top half, one in the bottom. Both faces stay visible the whole time, so reactions are never lost.

  • Works best for: debates, reactions, moments where the listener’s face is half the story.
  • Watch out for: captions. The seam between the two halves is the natural place for them — on a face is the worst.
  • Limit: it only works with two people who are each framed reasonably large in the source.

Option 3: Fit with a blurred background

Keep the entire wide frame, scaled to fit the width, and fill the space above and below with a blurred copy. Nothing is cut off.

  • Works best for: wide shots with three or more people, screen recordings, slides, anything where the whole frame carries information.
  • Watch out for: small faces. The picture becomes a strip in the middle of the phone, so this is the weakest option for a close, personal conversation.

Option 4: Center crop

A fixed crop in the middle of the frame. Simple and stable — and ideal when the source is already roughly vertical, or a single speaker sits centred in a wide shot.

Choosing quickly

  1. One person in the frame? Center or Follow speaker.
  2. Two people, conversation? Follow speaker.
  3. Two people, reactions matter? Split.
  4. Three or more people, or a screen? Fit.

Don’t forget the other shapes

Not every placement is 9:16. Square (1:1) and portrait (4:5) are common in feeds, and 16:9 is still right for regular YouTube uploads and many websites. A 4:5 crop keeps noticeably more of a wide frame than 9:16, which can rescue a shot where two people sit close together.

The same moment in 1:1. Footage: ITU · AI for Good · CC BY. Rendered by Klyp.

Captions and hooks follow the framing

Whatever layout you pick, captions and the hook title have to live around it. In a follow-speaker crop the face is usually in the upper-middle, so captions sit lower. In a split, they belong on the seam. Check every clip once with the captions on — the best framing in the world loses to a caption across someone’s mouth.

Written by The Klyp team

Published . Spotted a mistake? Tell us.

Turn your next long video into clips.

Klyp finds the moments, frames the speaker and writes the captions. Free to start, no card.

How it works