UGC ads are ads shot in the style of ordinary posts, usually by a creator you commission. TikTok’s own creative best practices recommend featuring creators, employees or customers and going for a DIY look that is “not overly polished”, so the format fits the feed it runs in.
After the offer, the format is the biggest lever a UGC buyer controls. Pick the wrong one and a good product reads as a sales pitch. This guide covers seven formats, a way to choose between them, and what to settle with a creator before they hit record.
Choose the format by what the viewer already knows#
Copywriter Eugene Schwartz described levels of buyer awareness in Breakthrough Advertising (1966). A short version works for UGC planning:
| Viewer knows… | They need | Start with |
|---|---|---|
| The problem, not the solution | To see the problem solved | Problem-solution demo, routine integration |
| That solutions exist, not yours | A reason to pick yours | Comparison, expert explainer |
| Your product, not why to buy now | Proof and risk removal | Honest review, unboxing, comment reply |
Run one format per audience stage in your first test. Mixing formats and audiences in one ad set makes the result unreadable.
The seven formats#
1. Problem-solution demo#
- Use when the product fixes something visible and the buyer already feels the pain.
- Beats (20 to 30 seconds): problem on camera (0 to 3s), product enters and is used (3 to 15s), result shown (15 to 25s), one line of next step.
- Brief line: “Open on the problem happening for real. Show the product in use before you explain it. End on the result, close up.”
- Watch for staged-looking problems. Messy is believable.
2. Honest review#
- Use when the buyer is product-aware and comparing, or the price needs justifying.
- Beats (30 to 45s): verdict first (0 to 3s), what you expected, what happened, one downside, who it is for.
- Brief line: “Give your verdict in the first sentence. Include one real downside. Say who should not buy it.”
- Watch for reviews with no downside. Viewers read them as paid.
3. Unboxing and first impressions#
- Use when the packaging, scale or contents are part of the appeal: gifting, bundles, subscription kits.
- Beats (20 to 40s): package arrival, one reaction line, three details the creator notices, first use.
- Brief line: “Film the first time you open it. Say what surprises you about size, feel or smell.”
- Watch for long openings. Start at the reveal, not at the courier.
4. Routine integration#
- Use when the product lives inside a habit: skincare, supplements, cooking, pet care, cleaning.
- Beats (25 to 40s): situation (“my 7am routine”), product appears naturally, one detail on how it fits, outcome after the habit.
- Brief line: “Show the product as one step in a routine you already do. Mention it as you would a friend’s recommendation.”
- Watch for a routine so busy the product disappears. Keep it in frame for at least five seconds.
5. Comparison: “I tried three”#
- Use when the category is crowded and the buyer is solution-aware.
- Beats (30 to 45s): the test setup (0 to 5s), each option under the same conditions, a ranking, why yours won.
- Brief line: “Test all options the same way, on camera. Do not name a competitor negatively. Show the measurement.”
- Watch for rigged tests. Run it for real, and keep the footage.
6. Comment reply#
- Use when you have an objection that keeps appearing in comments: “does it leave residue?”, “is it loud?”
- Beats (15 to 25s): the comment on screen (0 to 2s), direct answer, demonstration, short line to close.
- Brief line: “Reply to this exact comment: [paste]. Answer in one sentence, then prove it.”
- Watch for invented comments. Reply to real ones.
7. Expert explainer#
- Use when the product claims a mechanism the buyer does not understand: ingredients, materials, technique.
- Beats (30 to 50s): credential and claim (0 to 5s), explanation with a simple visual, one limitation, recommendation.
- Brief line: “State your real experience in the first line. Explain the mechanism in one visual. Name one thing the product cannot do.”
- Watch for credentials that go beyond what the creator holds, and for claims that need substantiation.
Where to find the formats that work in your category#
Before briefing, check which formats dominate the strongest videos in your niche. That means the same exercise as in the hook research guide, applied to format. Pull a recent set across TikTok, Reels and Shorts, normalize views against each creator’s median, and count formats in the top quartile against the whole set. In ContHunt, a Deep Search prompt does this in one pass:
For kitchen gadgets in the last 30 days, label each video’s format (problem-solution demo, honest review, unboxing, routine integration, comparison, comment reply, expert explainer, other). Show the format mix of the top quartile versus all videos, with example links.
What to settle before the creator records#
Usage rights. Write down where the footage may run: organic repost, paid ads, how long, which countries. For paid use on TikTok, Spark Ads can promote another creator’s post only with that creator’s authorization, so ask for it in the agreement.
Disclosure. The FTC’s guidance for social media influencers says the disclosure belongs in the video itself, not only in the description, and works best in both audio and video. Plain terms such as “ad” or “sponsored” are acceptable. Vague tags like “collab” are not.
Claims. The creator may only say what you can substantiate. List forbidden claims in the brief, especially for health, skin and money products.
Real people only. The FTC’s final rule on fake reviews and testimonials covers AI-generated fake reviews. Do not present synthetic “customers” as real ones.
Deliverables. Ask for 9:16 files, raw footage, and three alternative openings on one body.
Test in this order#
- Hook. Same body, three openings. This is the cheapest test and usually the most informative.
- Format. Same offer and audience, different format from the list above.
- Offer and call to action. Only after hook and format settle.
Change one variable per test, and write the hypothesis before launch. A test with no hypothesis teaches nothing when it loses.
Brief template you can reuse for any format#
FORMAT: [one of the seven]AUDIENCE STAGE: [problem-aware / solution-aware / product-aware]JOB: [Audience] should [believe/do] because they see [proof].BEATS: [timed beats from the format above]OPENING VARIANTS: A / B / CMUST SHOW: [product use, proof moment]MUST NOT SAY: [claims list]RIGHTS: [platforms, paid use, duration, authorization]DISCLOSURE: in video, audio and textDELIVER: 9:16, raw footage, 3 openingsGet started with a Deep Search on your category, find which formats lead it, and fill this template with videos you can point to.
