Two Ways to Organise Source Files Before Editing Starts

Why the filing method matters before any editing

Most problems that surface late in an audio edit were planted in the first hour, when files were copied across and given names in a hurry. A recording called take_final_2 tells you nothing three days later, and a producer who returns to a project after a week should not need to listen to every clip to find the right one. This article compares two sensible ways of organising authorised recordings alongside an approved script, so that the decision is made once and made deliberately.

Both methods assume the same starting point: you hold recordings you are permitted to use, and you have a script that the client or commissioning editor has signed off. Nothing is edited until those two things are confirmed. The question is only how the material is laid out around them. The two options below are not rivals in any dramatic sense, but they behave differently as a project grows, and that difference is worth understanding before you commit a folder structure to a real job.

Option one: folders grouped by speaker

The first method sorts everything by who is talking. Each narrator, interviewee or reporter gets a folder, and inside it the takes are named in the order they were recorded, with a short note on the microphone or room if it changed. This suits interview films and recorded reports well, because the cleanup work that follows is mostly per voice: a hum that affects one guest, a harsh sibilance on one presenter, a different room tone for the person who dialled in from an office.

The strength of this approach is that processing decisions stay together. When you correct one speaker, every clip you need is in one place, and you can apply the same settings consistently without hunting. The weakness appears when you try to assemble the programme. The script runs in a sequence that cuts between voices, so you spend time jumping between folders to find the line that belongs at a given point, and gaps in coverage are harder to spot at a glance.

• Keeps cleanup settings consistent for each voice
• Works well for multi-guest interviews and panel recordings
• Makes script gaps harder to notice

Option two: folders grouped by script section

The second method follows the approved script instead. Each section, scene or paragraph of the script becomes a folder, and every take that attempts that passage is filed beneath it. For a single-narrator explainer or a pre-recorded programme segment this mirrors the way the edit will actually be built, since you move down the script line by line and choose the best delivery of each part.

Here the strength is traceability. You can open the script, point to any paragraph, and know exactly which takes exist for it and which are missing. Reviewers who only see the script can follow your decisions without learning a naming scheme. The cost is that speaker-level processing is scattered. If one voice needs noise reduction, you now apply it across many folders, and it is easier to treat the same person slightly differently in two sections without noticing until the levels are balanced later.

• Mirrors the order in which the edit is assembled
• Missing passages show up immediately against the script
• Per-speaker processing is spread across several folders

Naming rules that work with either method

Whichever structure you choose, the file names carry most of the weight. A reliable pattern puts the project code first, then the speaker initials, then the script section number, then the take number, separated by underscores and with no spaces. Add a date only if takes were recorded on different days. Keep the pattern short enough to read in a narrow timeline lane, because long names are cut off exactly where you need them.

Never rename files after you have started editing unless your software relinks them reliably, and keep an untouched copy of the originals in a separate folder marked as source. That copy is your safety net when a clip turns out to have been trimmed too tightly. A plain text note beside the folders, listing the script version, the date it was approved and who approved it, saves a surprising amount of argument later about which wording was final.

• Project code, speaker initials, section number, take number
• Original files kept untouched in a separate source folder
• A short note recording the approved script version

Which one we recommend, and when to break the rule

For most of the work our learners bring, we recommend grouping by script section and recording the speaker in the file name. A recorded report, an explainer or a programme segment is judged against its script, and the producer who reviews it thinks in script order. Having every take for a passage in one place makes choosing between deliveries quick, and it exposes missing lines before they become a reshoot request.

The exception is the long interview film with several guests and little scripted narration. There the speaker folders earn their place, because the real labour is cleanup and the voices differ more than the passages do. In that case keep speaker folders, but add a simple index sheet that maps each guest's best takes to the cut. If you are unsure, start with the script-section layout; converting it later is easier than untangling a speaker layout that never matched the script.

A short check before you open the timeline

Before you edit anything, run through a few plain questions. Do you have written permission to use every recording? Is the script the approved version, and is its version recorded somewhere? Does every file open and play to the end? Are the original files copied and set aside untouched? Each of these takes a minute, and a no to any of them is a reason to stop and ask rather than carry on.

This is the habit we spend the first module of the course building, because it changes how the rest of the work feels. When the material is labelled and the paperwork is settled, the editing itself becomes calmer: you are choosing between known takes rather than searching for them. The time you invest in organisation is returned many times over during cleanup, balancing and review, when a clear structure means feedback is applied to the right file the first time.