# UGC Facebook Ads That Keep Running: Open on a Person, Then Earn the Offer > In one search of the Marxx library, 30 of the 50 longest-running Facebook video ads opened on a person's story, and only one on an offer. - **Author**: Marxx - **Published**: 2026-11-02 - **Category**: Production - **URL**: https://marxx.ai/posts/ugc-facebook-ads-that-keep-running --- **How do you make a UGC Facebook ad that people keep watching?** Start with a person, not a product. In one pull from the Marxx library, 30 of the 50 longest-running Facebook video ads we read opened on someone telling a story: a customer, a founder or a creator. Only one opened on an offer. # UGC Facebook Ads That Keep Running: Open on a Person, Then Earn the Offer Most guides to UGC ads tell you to film a real customer, add captions, and push the discount. That's a reasonable recipe, but it's untested against what actually keeps running. So we looked at what keeps running. We pulled 500 active Facebook video ads that have run 90 days or more from the Marxx library, across all industries, and read the top 50. For each one we coded the opening from its hook field: a person-led story, a question or pain, a product or claim, a brand or recommendation, or an offer. The counts are approximate and come from one search. Most long-running Facebook video ads open on a person: 30 story, 11 question or pain, 5 product or claim, 3 brand or recommendation, 1 offer ## Story openings are the most common Thirty of the 50 open on a person telling a story. The people vary: a customer describing a result, a founder explaining why the brand exists, a creator reacting to a problem. What they share is that the first line is about their experience, not the product. That matters for UGC specifically. A UGC ad isn't just a customer on camera. It's a customer whose first sentence is the story the viewer recognises, and the product appears inside it. The sample doesn't tell us which of those openings causes the run to last, only that they are the common opening among the longer runners. ## Questions and pain points run longer Eleven of the 50 open on a question or a pain point. Those ads have the longest median run in the sample, about 165 days, against about 131 for the story openings. That's a small group, so treat the gap as a reason to test, not a rule. Stories were most common; questions ran longer. Median days live by opening type, with sample sizes The longest-running question-led ad in the sample is Grow Digitally's "Are you a coach or a digital marketer?", which has run 232 days. It names the viewer's situation in the first line, then offers a fix. ## Offers come later One of the 50 opens on an offer. That's the clearest pattern in the sample. Long-running UGC ads earn the offer after the viewer has been drawn in, not before. This is a different pattern from a broader sample of Facebook video ads. In an earlier pull across offer-led video ads in the same library, offer-first openings were more common. The two searches used different queries, so they can't be compared directly. The lesson is narrower: when the ad is UGC, the sample favours a person first. ## Four openings that kept running These are verbatim hook fields from the library, with the days each ad has been live. Four openings that kept running: two stories and two questions, with verbatim hooks and days live The two stories are a seller describing a shirt turned into a photoshoot, and a buyer describing weeks of trying to sell a car alone. The two questions do similar work. One names the viewer's job, the other offers a result the viewer wants without requiring them to show their face. ## A short UGC test you can run Don't copy these openings. Test the pattern on your own product. 1. **Write three openings for one offer:** a customer story, a question about the viewer's situation, and an offer-first line. 2. **Film the story opening with a real customer,** not an actor, and keep it under 30 seconds to the first claim. 3. **Run the three as separate ads** with the same audience and budget. 4. **Read day 14 and day 30,** not day one. Keep the opening that still runs. The test is cheap because the three versions differ only in the first line. The rest of the ad can stay the same. ## What this sample cannot tell you This is one search of the top 50 of 500 results, and only ads still active when we pulled them. Ads that stopped before 90 days aren't in the sample, so this describes what survived, not what brands tried. The hook coding is ours, done by hand, and the counts are approximate. Some ads in the sample are not UGC in the strict sense, such as founder-led or creator-led ads, so the category is broader than "customers filmed on a phone". Medians come from small groups, so read them as direction. ## Where Marxx fits The Marxx generation canvas builds UGC-style video from one brief: script, voice, avatar or UGC talent, and resizing into Facebook placements. The useful part for this test is writing the three openings from one brief, so each version differs only in the first line. **[Trial the canvas: turn one brief into a finished ad ->](https://www.marxx.ai)** *Related: [Best AI UGC tools](/blog/best-ai-ugc-tools) - [Automatic reel makers](/blog/automatic-reel-makers) - [Facebook video ad makers compared](/blog/facebook-video-ad-makers)* ## FAQ **Do UGC ads need a real customer?** Not always. The sample includes founder-led and creator-led ads as well as customers. What they share is a first line about a person's experience. Test the opening before you commit to a casting. **How long should the opening be?** Keep the person's story or question inside the first few seconds, then move to the product. We haven't measured an exact length in this sample, so test it on your own ads. **Is an offer ever the right first line?** It can be. In this sample it's rare, so test an offer-first version against a story opening rather than assuming it wins. --- - [More Production articles](https://marxx.ai/posts/category/production) - [All articles](https://marxx.ai/posts)