Introduction — The Quick Clean-Up That Took Eleven Hours
A ninety-minute quote became eleven hours on one merged track.
Wren quoted a quick clean-up on one episode at about an hour and a half.
It took eleven.
The episode was seventy-two minutes long. The host and the guest had been recorded on a single merged track, because the host had opened a laptop, pressed record on the built-in microphone, and put the guest on speakerphone in the middle of the table. The room had a refrigerator in it. The guest's line dropped four times, and each time the host repeated the question and the guest re-answered it, so four sections existed twice with slightly different wording. Forty minutes in, the conversation went somewhere the host later described as "a bit of a tangent, can we lose that", and the tangent turned out to be structural rather than decorative — the guest referred back to it eleven times afterwards, so removing it meant finding eleven places where a sentence no longer made sense and building a bridge over each one.
Then there were the things nobody had mentioned. An intro. Show notes. Timestamps. Three clips for social. A title, a description and the episode uploaded to the host's platform. None of that had been discussed, because the host had asked for a clean-up and genuinely believed a clean-up was one thing.
Eleven hours, at a price set for ninety minutes.
Nobody behaved badly in that job. The host was not trying to get more than they had paid for. They did not know that one merged track cannot be separately processed, that a refrigerator is permanent, that removing forty minutes from the middle of a conversation is not a deletion but a rebuild, or that show notes are a piece of writing rather than a formatting task. They knew that the finished episodes they listened to sounded good, and they assumed the difference between their recording and those was a button.
That job would be a story about an inexperienced editor if it had ended there. It did not.
Month four. Wren placed a music bed under the intro, taken from a download described as royalty-free. Nine weeks later the show received a claim on that music, and the host paid to resolve it. The track was not free for this use, in this territory, on a monetised show. Wren had not read the licence, because there had not been a licence — only a description. The episode had been downloaded by several thousand people, and every one of those copies still contained it.
Month seven. A guest wrote, three weeks after publication, asking for a section to be removed. They had told a story that identified their employer, and they had thought better of it. Wren cut the section, re-uploaded the file, and told the guest it was done. It was not done. Every application that had already downloaded the episode still had the original, the section had been quoted in a clip that was still circulating, and the transcript was indexed. The guest believed the material was gone. Wren had allowed them to believe it.
Month nine. An edit removed a hesitation from a sentence about a named company. The hesitation had come immediately before the word "allegedly". What went into the finished cut was a flat factual assertion that the company had done something. The host caught it on their listen-through, with two hours to spare. If they had been busy that week, it would have gone out.
Month twelve. The show went from weekly to twice weekly, and asked Wren to keep going. The price was still the one set on that first eleven-hour episode, agreed before anybody knew what the work was.
None of those four were audio problems. Wren's editing had been technically competent throughout. What failed was procedure — and in this trade, three of the four failures are unrecoverable once they are in a feed.
This book is built on the four properties of the work that produced them.
The recording already happened and you cannot make it again. What lands in your folder is an unrepeatable event with its faults baked in. You will be asked to fix things that cannot be fixed, and the useful skill is knowing which is which before you quote.
Editing is deciding what somebody said. Every cut is an editorial judgement on somebody else's words, published under their name. Most of those judgements are trivial. A small number of them are the difference between a tidy sentence and a false statement about a person who is not in the room.
Published is distributed and cannot be recalled. A feed is a push mechanism. Replacing a file does not reach the copies already downloaded, does not retract a claim, and does not restore a guest's consent. There is no version of "we took it down" that is fully true.
You are paid per episode and you are hired for the season. The trial episode sets a price that will still be in force two years later, across a schedule that can double, on source material whose condition varies every week while the fee does not.
The book runs on two numbers.
Revenue per production hour — net revenue divided by every hour from receiving files to publication. Not per episode, because a seventy-two-minute interview from two clean tracks and a seventy-two-minute interview from one merged track are the same line on an invoice and a fourfold difference in work.
Finished minutes per production hour — total delivered minutes divided by hours worked. It costs a note in a log. It responds within one episode to whether you insisted on separate tracks and used a session template. And almost nobody in this trade keeps it, which is why almost nobody in this trade can explain why some months pay and some do not.
By month twenty-six, Wren was at roughly 34 finished minutes per production hour, up from about 9. Episodes published without a correction had gone from around 74% to around 98%. Revenue per production hour had roughly tripled, and the base rate had not moved in the first year. The change was entirely in what happened before the first cut.
The revenue mix by then: retained weekly production 38%, full show management 22%, clip and short-form packages 16%, one-off episode edits 12%, launch and setup work 12%.
Read Chapters 5 to 8 first, in order. They are the four rules, and every later chapter refers back to them.
Then read Chapter 20 before your next intake, because the source files decide more about a job's profitability than any decision you will make afterwards.
If you want the single highest-return action available today: ask for separate tracks, and write down the deliverable list one item at a time before you give a price. That is the whole difference between ninety minutes and eleven hours.
©2026 James Henderson / https://localhandyman.work