Verified against ChatGPT · 2026-08-09
Write a UGC creator brief that gets an authentic-feeling video without scripting away the authenticity
Produces a brief for a UGC creator making a short product video — talking points and required beats, not a word-for-word script — tuned so the result still feels like a real person's honest take.
The prompt
Ready to copy — highlighted parts are example details you can swap.
Write a brief for a UGC creator who will film a short product video for paid social. This is a talking-points brief, not a word-for-word script — a creator reading a script on camera is the exact failure mode that makes UGC stop feeling authentic and start looking like an ad, which defeats the point of hiring a creator instead of doing a studio ad. PRODUCT An overnight retinol serum for first-time retinol users. PLATFORM AND FORMAT TikTok/Reels, 30-45 seconds, vertical, native-feeling not polished. MUST-HIT POINTS (the only non-negotiable content) Start with the lowest concentration (0.25%) for first-time users, apply only at night, always follow with SPF the next morning. CREATOR'S OWN CONTEXT The creator has documented sensitive, acne-prone skin on their channel before and has an audience that specifically asks about gentle skincare. THINGS THAT MUST NOT BE CLAIMED Cannot say it 'cures' or 'reverses' aging, cannot claim visible results in a specific number of days without a substantiated study behind that claim. STRUCTURE (beats, not lines) Give the creator a beat structure: hook (what problem or moment opens the video), the honest reaction or demo moment, the must-hit points woven in as things the creator would naturally say, and a close. For each beat, describe what should happen and why, in enough detail that a creator understands the intent — but do not write dialogue for them to read verbatim; leave the actual words to the creator's own voice. USE THE CREATOR'S OWN CONTEXT Work the creator's own stated context into the hook or reaction beat specifically — a UGC brief that ignores who this particular creator is and could apply to any creator is indistinguishable from a script, which is the thing being avoided. COMPLIANCE GUARDRAILS State the things that must not be claimed as a plain do-not-say list, and briefly explain why each matters (a specific regulatory, medical, or overstatement risk) so the creator understands the boundary rather than just following a rule blindly — a creator who understands why a claim is off-limits is less likely to accidentally rephrase into something equally risky. WHAT NOT TO DO Do not write more than one example line per beat, and label any example line clearly as "example phrasing only, say it your way" — never present example lines as the required script. OUTPUT FORMAT Beat-by-beat structure with intent notes, the do-not-say list with reasons, and any example phrasing clearly labeled as optional.
Customize
Optional — swap in your own details for the highlighted parts above.
Why this works
The explicit prohibition on writing dialogue counters GPT-5.1's strong default when asked for a 'video brief' — without that constraint, it tends to produce a beat structure that's actually a disguised script, complete with specific sentences meant to be delivered close to verbatim, because that's the easiest way to guarantee the must-hit points get said and it mirrors the shape of most ad-script examples in its training data; the problem is that a UGC creator reading assigned lines produces the flat, over-articulated delivery audiences have learned to recognize and scroll past, which is precisely the outcome paying for 'user-generated' content is meant to avoid. Requiring the creator's own stated context to be worked into a specific beat, rather than left as unused background, forces the brief to actually be creator-specific rather than a generic template that happens to name this creator in the header — a brief that would read identically for any creator given the same product has failed at the one thing that makes UGC different from a script read by an actor. Pairing the do-not-say list with the reason behind each restriction, rather than a bare rule list, matters because a creator who understands that 'reverses aging' is a legally risky claim (not just an arbitrary brand preference) is far less likely to accidentally say something equally non-compliant in their own words during an ad-libbed take — a bare prohibition list only blocks the exact phrase given, while an understood reason generalizes to phrasing the brief-writer never anticipated.
What you get back
Hook beat: open on the creator's own documented sensitive-skin story — "you've talked about your skin freaking out before, start there and show why you were nervous to try retinol at all." Example phrasing (optional, say it your way): "I was so scared retinol would wreck my skin like everything else has."
Verified against
ChatGPT GPT-5.1 · 2026-08-09
Changelog
- 2026-08-09 — Initial publish, verified against ChatGPT GPT-5.1.
Need this built into your business?
If a prompt isn't enough — Google Ads management, built and maintained for you — that's Scult's day job.
EXPLORE GOOGLE ADS MANAGEMENT
