How to Cut Down a Long Interview Without Losing the Story

Somewhere on a hard drive right now is a sixty-minute interview that only needs to run for ninety seconds, and the editor staring at it has already watched the opening six minutes four times. (The good line is always at minute forty-one. Always.) You cut down a long interview by choosing the handful of moments strong enough to stand alone and deleting everything else, not by shaving a little time off every section. That single shift in approach, picking instead of trimming, is the difference between a piece that feels inevitable and one that feels like it got smaller without getting better.

We've all stared at a timeline wondering why a cut that's technically shorter still feels bloated. Usually it's because the cutting happened in the wrong order: trim first, decide later. Flip that order and the editing software stage turns into assembly work instead of the place where you're still figuring out what the piece actually is.

Define the target

What does this cut actually need to do?

Write one sentence describing what this piece is and roughly how long it should run before you touch a single clip. A ninety-second social cutdown and a five-minute case study pull almost none of the same material from the same sixty-minute conversation, even when it's the same person talking about the same event on the same afternoon. Skip this step and every decision becomes a negotiation with yourself, because a good subject says a dozen genuinely interesting things an hour and nearly all of them are individually defensible to keep. The one-sentence brief is what lets you say no quickly, later, at 11pm, without relitigating the choice.

This is also where you decide what kind of cut you're making, because the filter changes with the format. A news package, a short soundbite reel, a full radio cut for a podcast episode, and a three-minute testimonial all start from the same raw footage but need a different lens applied from the first pass.

Read before you watch

Why should you build selects from the transcript instead of the scrub bar?

Reading the whole interview as text before touching footage lets you hold an entire hour-long conversation in view at once, instead of a ten-second window on a scrub bar. This is the backbone of what documentary and radio editors call a paper edit, and it predates every piece of modern software: editors were doing this with scissors and transcript pages long before anyone built a timeline for it. Public radio's Ira Glass has described the instinct for knowing the tape he needed the moment he heard it in an interview with the Columbia Journalism Review, a feel for the whole conversation that's much easier to develop on the page than in a scrub bar.

The catch: a transcript is only as useful as its timecode. If highlighting a line on the page doesn't hand you an exact in and out point on the actual recording, you're back to hunting for the moment by ear, which defeats the entire point of reading in the first place. That's the premise behind a transcript-based editor: every word is tied to a frame, so a highlighted sentence becomes a precise clip automatically, something you can confirm instead of something you have to hope is right.

The keep test

What actually earns a place in the final cut?

• Does it stand alone? A viewer who's heard nothing else in the piece should still follow the line without context you haven't given them yet.
• Does it earn its seconds? Every second on screen is a second of someone's attention. A merely-good line sitting next to a great one doesn't get a pass just for proximity.
• Does it serve the brief? Great material that's off-topic for this piece belongs in a different piece, not a compromise inside this one.
• Is it the cleanest version? Subjects often make the same point three different ways across an hour. Keep the sharpest delivery and let the other two go.

Running a full transcript through these four questions surfaces real candidates faster than any amount of scrubbing, because you're reading for ideas instead of waiting for them to arrive in real time. For a closer look at isolating the sharpest version of a given point once you've got rough selects, how to find the best soundbites in an interview covers that pass in more depth.

A worked example

How do you actually get from sixty minutes to three?

Picture a sixty-minute founder interview with a brief that boils down to a three-minute customer-facing piece about why the company exists. First pass: read the full transcript and mark every self-contained moment that could plausibly fit the brief, no trimming yet, just a rough highlight pass. A typical hour-long conversation usually yields twelve to twenty candidate passages this way, which is still far too many.

Second pass: run every highlight through the four-question test above and cut anything that fails even one. The founder's detailed answer about fundraising mechanics might be genuinely interesting, but against a brief about why the company exists, it fails the serves-the-brief test and gets dropped, no matter how sharp the delivery was. This usually cuts the candidate list roughly in half.

Third pass: trim the hedges and false starts inside each survivor, so a forty-five-second answer becomes a tight fifteen-second one without losing the point. Fourth pass: reorder. The strongest line, often something said twenty minutes in, becomes the open. The founder's actual first answer, usually a warm-up response about how the day is going, often doesn't survive at all. What's left should already read close to three minutes on the page, before a single clip has touched a timeline.

Cut inside the line

Why isn't picking the right moment enough on its own?

Selecting the right moment is only half the job, because most people don't speak in clean sentences. A strong idea often arrives wrapped in false starts and hedges, the kind of rambling warm-up where someone circles a point three times before landing it. Pulling the hedges and restarts out doesn't change what someone meant, it just removes the static around it, and the practical effect is that subjects sound like the sharpest version of themselves rather than a nervous one. Treat this as its own pass, separate from selecting the moment in the first place; how to remove filler words from an interview walks through exactly where that line sits.

Reorder for momentum

Does the interview have to play back in the order it was recorded?

No, and it almost never should. People answer questions in the order you asked them, not in the order that makes the best piece, so the strongest material usually belongs at the open, as a hook, or at the close, as a payoff, regardless of when it was actually said. Reordering complete answers and tightening the language inside them are both normal, honest editing choices every working editor makes. There's a hard line just past that, though: stitching words spoken at different times into a sentence the subject never actually said. That's a frankenbite, and it's fabrication, not editing, no matter how clean the splice sounds.

The real math

How far can a ratio like this actually go?

For The Fog of War (2003), Errol Morris recorded 20 hours of McNamara talking for a film that runs under two hours. That's a real interview-to-runtime ratio worth sitting with: twenty hours of raw conversation reduced to a film built almost entirely around the strongest self-contained passages of one man reflecting on the biggest decisions of his career, with everything else, however interesting, left on the floor. Morris is well known for resisting a written paper edit on his own projects, preferring to find structure directly from the footage. Even so, the underlying discipline, pulling the strongest material and discarding the rest without mercy, is the same discipline any editor applies at any scale, paper edit or not.

You don't need twenty hours of source to learn from that ratio. A sixty-minute founder interview headed for a five-minute piece follows the same shape: self-contained moments pulled first, repetition and tangents cut hard, survivors tightened and reordered, all before anyone opens a timeline.

Common mistakes

What keeps going wrong even for experienced editors?

• Trimming evenly. Cutting ten percent off everything just makes the same sprawling structure ten percent shorter. It isn't the same as choosing what matters, and the finished piece will still feel unfocused.
• Selecting on the timeline. Scrubbing through raw footage in an NLE trades the wide view of a transcript for a narrow, slow one, and you end up rewatching the same twenty minutes three times.
• Keeping tangents because they're good. A brilliant aside that doesn't serve the brief belongs in a different piece, not a compromise on this one. Save it.
• Over-tightening. Stripping every pause and breath makes a subject sound rushed instead of confident. Leave room for people to breathe between ideas.

The honest tradeoff

Is it worth cutting material that's genuinely good?

Yes, and it still won't feel great in the moment you're deleting it. Cutting hard means throwing away good material on purpose, and that's an uncomfortable habit to build. It's worth it anyway: attention is the budget you're spending, and a tight five-minute piece holds it far better than a rambling twelve-minute one, even when the twelve-minute version has more good lines buried somewhere inside it. None of this works, though, if the transcript is wrong in the first place. A line that reads well on the page but was mis-transcribed gets selected, trimmed, and placed before anyone notices the actual audio doesn't match, which is exactly why getting the transcription itself right is the first quality gate, not an afterthought.

Where the pre-edit ends

When should you actually open the NLE?

You open it once the piece is locked on the page, not before. Defining the target, reading for selects, tightening lines, and reordering for flow together make up the pre-edit, the work that happens before an editing application opens at all. We built ScriptCut around exactly that sequence: a word-level timecoded transcript, a place to mark picks with AI Select, a tool that can arrange them into a rough order automatically with AI Arrange, and Cleanup to pull the hedges and false starts out of a selected line without re-recording anyone. A client reviews and approves the structure on a share link before anyone opens a timeline, and the finished, frame-accurate cut lands in DaVinci Resolve, Premiere Pro, Final Cut Pro or Avid, so the first thing you see in your NLE is already a cut, not a pile of raw footage waiting to become one. If you want the fuller workflow this sits inside, how to edit an interview faster covers the transcript-first version of the whole process in more depth.

If the footage is also feeding a podcast or a clips program, the same selects double as source material: ScriptCut Social can pull the moments worth posting from the same long recording, frame them on the speaker, caption them, and write a platform-specific description for each one, all from one approval link.

Here's how we'd start: open the transcript, not the timeline, and write that one-sentence brief before anything else. Everything downstream gets easier once the destination is fixed. When you're ready to move past the page, ScriptCut is built to carry the picks you made straight into a frame-accurate assembly.

Sources

• Wikipedia: The Fog of War
• Errol Morris: Fog of War
• CounterPunch: The Collaboration of Robert S. McNamara and Errol Morris
• Columbia Journalism Review: Ira Glass on structuring stories