Direct Response Video Built on a Substantiation File

Nothing gets shot until the substantiation file is complete. That file is a numbered list, and every proof statement in the script carries the number of the line it comes from: the study, the internal data pull with the query attached, the customer's recorded words with a timecode, or the authorised wording from the EU health claims register. A ninety-second direct response script typically holds five to nine proof statements, and in the categories I work in about a third of them do not survive contact with the file. Finding that out on a Tuesday in pre-production costs a rewrite. Finding it out after the shoot costs the shoot. The form is the old one because it still works: problem, mechanism, proof, offer, and a reason the risk is not the viewer's. The mechanism section is where most brand-trained teams flinch, because explaining why the product works feels like the unglamorous part and it is the part that makes the proof land. On YouTube in-stream the skip button arrives at five seconds, so the problem has to be recognised inside those five seconds by the person who has it and ignored by everyone else. That is a targeting job done with the script rather than with the audience settings. The substantiation file: I build it with you before writing, and it stays with you afterwards as an asset your next agency will not have. It is what makes legal review fast, because your counsel reads claims against sources rather than reading a script and imagining what might be behind it. In health and wellness it is the difference between a review cycle of three days and one of three weeks. For law firms it is where jurisdiction-specific limits on outcome and success-rate language get recorded once, so the next campaign does not relitigate them. Production: One or two shoot days depending on how many proof segments need staging. Founder or expert on camera runs on a teleprompter with the script broken into short takes, because reading long-form to camera is a skill most founders do not have and there is no reason to spend a shoot day discovering it. Cameras record isolated angles throughout, so cutdowns come out of the same material rather than out of a second shoot. Sound is recorded properly on lavalier and boom together; more direct response video is undone by room tone than by picture. Cutdowns and the offer layer: The master runs 60 to 120 seconds. Cutdowns at 30, 15 and 6 seconds come from the same session and are cut as arguments in their own right, not as trimmed versions that stop mid-sentence. The offer stays on its own editable layer, delivered as a template your team can change: when the price moves or the promotion ends, you swap the overlay in an afternoon without a re-edit and, in a regulated category, without a second legal review of everything around it. Not included: Media buying and YouTube campaign setup. Landing page design and copy, although I will read the page before agreeing to the work and tell you if it is the constraint. Ongoing editing of variants after handover. Any claim your counsel has not cleared, at any point, for any reason. I turn this down when: Nothing is converting yet in any format. Direct response video amplifies a working argument and cannot manufacture one, and every euro of production spend on a proposition that has not been proven in cheaper media is a euro spent to learn something a static ad would have taught for a fraction of it. I also decline briefs that want a countdown timer, a fabricated scarcity number, an earnings claim, or before-and-after imagery without a documented protocol behind it. And I do not produce for an offer the founder cannot state in one sentence — if it takes a paragraph on a call, it will take three on camera.