UGC questions › Ads by industry

Does UGC work outside ecommerce?

TL;DR

Yes, UGC works for service businesses, but the ads look different. We studied 25,697 Meta video ads that were running across 54 markets. Home-service and clinic ads stay live longer than online-store ads. They ask people to call, book or get a quote instead of "shop now", and the business owner or real job footage shows up more often. Most UGC advice is written for product stores, so it misses these differences.

Key findings

  • 39% of home-service video ads running today are 90+ days old. For online stores it's 27%.
  • 42% of home-service ads ask people to call, book, get a quote, apply or send a message. Only 2% of store ads do.
  • 23% of home-service ads send people to Messenger, compared with 1% of store ads.
  • 25% of home-service ads offer a free quote or estimate, and 25% name a city or state.
  • In finance ads, the owner or an expert is on camera 45% of the time. In coaching ads it's 40%. In store ads it's 11%.
  • Home-service businesses run fewer ads at once. Only 2% have 10 or more live, compared with 6% of stores.
  • Some things don't change. Service ads and store ads use AI people, customer stories and guarantees about as often.

Why ask whether UGC works outside ecommerce?

TL;DR: Most UGC guides show online stores, so service businesses have to guess. We wanted real numbers for them.

Search for UGC examples and nearly every result shows a product: an unboxing, a skincare routine or a supplement review. The proof usually comes from product reviews on store pages. That leaves out a large group of advertisers, like roofers, dentists, mortgage lenders, agencies and coaches, who sell a service and can't ship a creator a sample.

We work with many of these businesses, and they ask us the same questions. Does a creator ad make sense for a plumber? Should the owner be on camera? What should the ad ask people to do? We used Meta's public ad data to answer them.

What data did we use?

TL;DR: We looked at 25,697 video ads that were live on Meta in late 2026, plus a smaller group we tracked from launch to see how long ads last.

DatasetSizeWhat it tells us
Live video ads across 54 market searches25,697 ads from 10,209 advertisersHow old live ads are, what they ask for, what the copy says
Long-running ads with transcripts, tagged by format1,509 adsWho is on camera and what format the ad uses
Launch cohorts (from our part 2 study)1,368 ads from 70 advertisersHow long ads run before they're switched off

All ads come from the Meta Ad Library, US delivery, pulled in October and November 2026. We grouped each market into one of eight industry groups. "Ecommerce" combines food and drink, supplements and wellness, and home, lifestyle and apparel stores.

We counted all video ads in each market, not only ads made by creators. Creator-style ads were 22% to 59% of the tagged long-running ads, depending on the industry. So these results describe how each industry uses video ads in general, with UGC as a large part of that.

Every comparison resamples advertisers, not ads, so one big advertiser can't swing a result. Each figure below carries its sample size, and our full tables show the confidence intervals.

Do service businesses keep their ads running longer?

TL;DR: Yes. Home-service and clinic ads that are running today are older than store ads. App ads are the youngest.

Industry groupLive adsAdvertisers90+ days old365+ days oldMedian age
Home services2,3241,37039%5.5%60 days
Dental, clinics and telehealth3,9081,99536%5.4%54 days
Travel and events1,68592630%5.2%32 days
Ecommerce (all three store groups)7,9282,88227%3.4%38 days
Coaching and courses3,0141,17426%3.5%37 days
Finance and lending2,12993025%2.6%31 days
B2B services and agencies1,77882225%2.6%28 days
Apps and SaaS2,9311,17421%1.8%31 days

Home services beat ecommerce by 12 points on ads 90+ days old (95% CI +8 to +16), and clinics by 9 points (+5 to +12). Finance, B2B and coaching look about the same as ecommerce. Apps sit 6 points lower.

This table shows the ads that are live now, so it can't show how many ads were switched off along the way. Our launch-cohort data can. There, 87% of local-service ads were still running at 90 days, against 64% for ecommerce, 49% for coaching and 43% for apps. The gap with apps and coaching is clear. The gap with ecommerce is too uncertain to call, because only 13 local-service advertisers were in that group.

What do service ads ask people to do?

TL;DR: Service ads ask for a call, a booking, a quote or a message. Store ads say "shop now".

This was the biggest and clearest difference in the data.

Call-to-action buttonHome servicesEcommerce
Call, book, get quote, apply or send message42%2%
Get quote (alone)25%0%
Shop now6%64%

Where the ad sends people differs too. 23% of home-service ads send people to Messenger, compared with 1% of store ads, and 58% send people to a website, compared with 89% of store ads. A homeowner with a leak wants to talk to someone. A shopper wants a product page.

For a UGC ad, that changes the script. The ending has to make calling or messaging feel easy and safe. It shouldn't push a product link. Our home services guide has script examples built around a quote or a call.

What do service ads say in the copy?

TL;DR: Service ads offer free quotes, name the town they serve and show a price more often than store ads.

What the ad copy includesHome servicesFinanceB2BCoachingEcommerce
Free quote, estimate, consult or audit25%0%
A named city or state25%5%
"Near you", "local" or "serving"18%4%
A dollar price27%37%35%34%10%

Blank cells are groups where we didn't compare that feature. The price result surprised us. Store ads, which sell products with fixed prices, show a price least often. Finance, B2B and coaching ads show one most, often as a monthly rate, a fee or the cost of a program.

The first line of the copy also differs. Service ads call out their audience far more often ("Homeowners in Dallas", "If you're a realtor"): 20% to 24% of finance, B2B, coaching and home-service ads open that way, against 7% of store ads.

Who is on camera in service ads?

TL;DR: In finance, coaching and B2B ads, the owner or an expert is often on camera. Stores and apps use creator-style people more.

Who is on camera (tagged long-running ads)FinanceCoachingB2BEcommerceApps
Founder or owner45%40%23%11%5%
Creator-style presenter33%40%35%59%56%

These figures come from a smaller set: 134 finance ads, 173 coaching ads and 431 ecommerce ads, each tagged by format. Our tags can't always tell an employee from a hired creator, so treat "creator-style" as an upper limit on hired creators.

Home-service and clinic ads also used real footage of the work far more: 36% of home-service and 37% of clinic long-runners showed real jobs or real patients, against 7% of store ads. Store ads leaned on product demos instead (31%, against 13% for home services).

For a service business, the owner, a real crew and before-and-after job footage may matter more than casting a creator who has never met you. Our guide on founder-led ads versus UGC covers when each works. A good UGC ad for a service business can mix them, with a presenter explaining the offer over real job footage.

Do service businesses run fewer ads?

TL;DR: Yes. Most businesses in every industry had just one live video ad. Very few home-service businesses had 10 or more.

Industry groupAdvertisersShare with 10+ live adsTop 10% run at least
Home services1,3102%3 ads
Dental, clinics and telehealth1,7983%4 ads
Ecommerce2,7466%6 ads

Finance, B2B, coaching and apps ran about as many ads as stores. These counts come from keyword searches, so they're a floor on what each business runs, and the gap is about direction, not exact size.

Fewer ads, kept on longer, is the pattern for local services. It may mean their ads keep working in a small local audience, or it may mean a busy owner doesn't refresh them. The data can't tell which. Either way, a local business doesn't need the 50-ad testing plans written for national stores. Our guide to how many creatives to test covers sizing a test to your budget.

What doesn't change outside ecommerce?

TL;DR: Some things are the same. Service ads and store ads use AI people, customer stories and guarantees about as often.

  • AI-generated people: 7% of store ads, 8% of home-service ads, 10% of clinic ads and 6% of B2B ads. None of these gaps is clear.
  • Testimonial or case-study format: 14% of store ads, 16% of clinic ads and 15% of B2B ads. Home services was lower (7%), but on a small sample.
  • Guarantees in the copy: not more common in service ads. 14% of store ads mention one, against 7% of home-service ads.
  • Inside local services: no single cut, like video length, copy length or button type, predicted which local ads lasted longer.

Testimonials from real customers carry extra rules. The FTC's rule on reviews and testimonials bans testimonials that misstate who the person is or what happened to them. That applies to a roofer's ad as much as a skincare brand's.

What should a service business do with this?

TL;DR: Build UGC around a call or a quote, put your owner or real work on screen, name your area, and keep your winners running.

  1. Match the ask to the buyer. End with a call, quote, booking or message, and make sure someone answers fast.
  2. Show where you work. Name the city or area in the copy and on screen.
  3. Put real proof on camera. Use the owner, a technician, real job footage or real patients (with consent). Use a presenter to explain the offer, not to pretend to be a customer.
  4. Lead with a clear offer. A free quote, a free consult or a stated price is common in long-running service ads.
  5. Run fewer ads, longer. Test a few strong concepts and keep the winners on. Compare your ads with others in your industry, not with store benchmarks. Our industry guide and pages for clinics, coaches and financial services and SaaS and apps go deeper.

What are the limits of this study?

TL;DR: Most of our data shows ads that are live now, not ads that were stopped. Some format results rest on small, tagged samples.

  • Live ads only. The 25,697-ad dataset holds ads running when we pulled them, so it shows what advertisers keep on. It can't give survival odds. Only the 1,368-ad cohort includes stopped ads.
  • Industry labels come from market searches. A "home services" search includes nearby and competing advertisers, and some advertisers appear in more than one group.
  • Format tags are on a selected subset. The 1,509 tagged ads were chosen as long-runners and references, and they're older in some groups than others. Tags were assigned by a model, with no second coder.
  • Copy checks use simple word lists. They treat every group the same, so the comparisons are fair, but the counts aren't exact.
  • Many comparisons. We ran several hundred comparisons, so a few will look different by chance. The findings above are the large gaps in large groups.
  • Run length isn't results. Meta shows no spend or sales for US commercial ads (Meta Business Help Center), so a long run means the advertiser kept paying, not that the ad was profitable. Mockingbird's guide explains what start dates can and can't tell you.

Sources

On Google

See more of She Converts in your search results

Add us as a preferred source and Google shows our UGC ad research more often in Top Stories and AI answers.

Add as preferred source

Want to see what this looks like for your ads?

Book a 30-minute call. We review your current ads and your offer and see if we're a fit. No retainer required.

Book a Call