Introduction — The Light Kit That Made Everything Worse
Four hundred dollars of lighting made every subsequent video perform worse.
Ash spent about four hundred dollars in month eight and made every subsequent video perform worse.
A light, a stand, a diffuser, a clip-on microphone, a small tripod. Everything a sensible person buys when they decide to take something seriously.
⚠ The videos got cleaner, better lit, better framed and better sounding. Two brands stopped rebooking within six weeks, and one of them said why: "The old ones looked like our customers. These look like an ad."
That sentence is the whole trade.
⚠ You are not being paid to make good video. You are being paid to make video that does not look bought — and every instinct you have about improving your work is pushing in exactly the wrong direction.
⚠ There is a floor. Inaudible sound, a dark room, a shaking hand, a product you cannot see — those read as bad rather than as real, and they fail too. ⚠ The window between "obviously homemade" and "obviously produced" is narrow, and living inside it deliberately is the actual skill in this business.
One video is not a deliverable
The second thing Ash had wrong was arithmetic.
Month five. Ash delivered a single, carefully made thirty-second video per brief — scripted, filmed, edited, colour-corrected, captioned. ⚠ About six hours of work each, and the brand's reply was: "This is lovely, but we can't test with one."
⚠ A brand running paid social is not commissioning a film. They are running a test — and a test needs variants. The same demonstration with three different opening lines. The same testimonial with two different endings. A vertical cut and a square one.
⚠ What the brand actually wanted from that fee was six or eight usable variations, and Ash had spent six hours making one very good one.
At month five Ash was producing about 0.9 variations per production hour. ⚠ By month twenty-four it was about 2.4 — with worse lighting, less editing and considerably less effort per clip.
The video that ran for fourteen months
Month ten, and the expensive one.
Ash filmed three clips for a skincare brand. Flat fee, no paperwork beyond an email, delivered and paid.
⚠ Fourteen months later one of them was still running as a paid advertisement. Not on the brand's account — on Ash's. The brand had been given access to run ads from Ash's own handle, which Ash had agreed to in a message without understanding what it meant.
⚠ By then Ash had signed with a competing skincare brand, who found the ad and were reasonably annoyed.
Nothing improper had happened. ⚠ No term had been agreed, no territory, no media, nothing about paid amplification, and nothing about category exclusivity. The brand did what nobody had said they could not do.
⚠ That is Rule 2, and it is the one this trade is worst at: you are not selling a video. You are licensing your face, in a use, for a period, on a platform, with or without money behind it — and if you name none of those things, you have granted all of them.
The sentence Ash said on camera
Month twelve, and the serious one.
A supplement brand supplied a script. ⚠ It contained a sentence about how the reader would feel after two weeks. Ash read it, filmed it, delivered it, and was paid.
⚠ The advertisement was later pulled. Ash's face was on it. Ash had never taken the product.
⚠ Nothing here is legal advice, and what responsibility sits where varies by country and by category — which is exactly the point. Ash had never asked, had no written position, and had said a sentence on camera about a product they had not used, in a category where that matters.
⚠ That is Rule 3, and it is different from every other kind of media work: a voice actor can be anonymous, an editor is invisible, a photographer is behind the camera. You are in front of it, saying "I tried this," with your name underneath.
What changed
The rebuild took about four months and almost none of it was filming.
⚠ Ash defined an authenticity floor and ceiling in writing — phone, window light, real room, no colour grade, sound checked every time, deliberate imperfection kept rather than removed. Chapter 5.
⚠ Ash started pricing per variation rather than per video. A concept became one setup, three hooks and two endings — six variations — and the quote said so. Chapter 8.
⚠ Ash wrote a likeness licence naming five things: term, territory, media, paid amplification and category exclusivity — plus a clause about synthesis and cloning. Chapter 6.
⚠ Ash wrote a claims policy: a list of sentences that will not be said on camera, a rule about actually using a product for a stated period before filming it, and a redline process for brand-supplied scripts. Chapter 7.
⚠ And Ash started asking every brand one uncomfortable question at thirty days: how many of these did you actually run?
The numbers, twenty-four months in
⚠ The shipped rate — the share of delivered variations a brand actually ran — went from about nineteen per cent to about fifty-four. That number is not on any dashboard; it exists only because Ash asked.
Variations per production hour: about 0.9 to about 2.4.
And the revenue mix at month twenty-four:
- Variation packs — 34%
- Monthly retainers — 24%
- ⚠ Usage extensions and whitelisting fees — 19%
- Scripting and concepting without filming — 13%
- Editing for other creators — 10%
⚠ The third line is the one that did not exist in year one. It is revenue from footage already delivered, arriving because somebody wrote down when a licence ended — and it costs no filming, no product and no time in front of a camera.
What this book is
Thirty-seven chapters, sixty-eight resources and one hundred and forty-eight prompts, organised around four rules and one licence.
⚠ It is not a filmmaking book. There is comparatively little here about craft, and what there is mostly tells you to do less. The decisions that changed Ash's business were about variation counts, licensing, claims and knowing which of two hundred delivered clips actually ran.
⚠ And two chapters matter more than the others in different ways. Rules 1 and 4 decide whether the business is profitable. Rules 2 and 3 decide whether it is safe — because one governs your face and the other governs your name, and both of those are attached to you permanently in a way that no file is.
Read Chapters 5 to 8 first, in order. Then read Chapters 6 and 7 again before you agree to anything.
©2026 James Henderson / https://localhandyman.work