A local Facebook video ad works best when it states the offer in the first seconds, uses on-screen captions, targets a nearby audience and asks for one action, such as a call or a booking. Pew Research Center's 2026 survey found that 70% of US adults use Facebook, so the audience is large. A large audience does not guarantee cheap or effective clicks, so plan a fixed test budget, run it for several weeks and compare cost per result before scaling. Worked examples below are labelled as assumptions.
↓
US adults on Facebook
Pew 2026, was 71% in 2025
→
US adults on YouTube
Pew 2026, unchanged from 2025
↑
US adults on Instagram
Pew 2026, up from 50% in 2025
→
US adults on TikTok
Pew 2026, unchanged from 2025
See Sources at the end of the article. Cards marked Assumption are examples, not results.
Pew's 2026 fact sheet describes YouTube and Facebook as the most widely used social media platforms in the United States, with about half of adults using Instagram. The survey covered 5,511 US adults between January 30 and June 18, 2026, with a margin of error of plus or minus 1.8 percentage points.
| Platform | 2025 share of US adults | 2026 share of US adults | Change |
|---|
| YouTube | 84% | 84% | No change |
| Facebook | 71% | 70% | Down 1 point |
| Instagram | 50% | 51% | Up 1 point |
| TikTok | 37% | 37% | No change |
| Reddit | 26% | 25% | Down 1 point |
| Snapchat | 25% | 26% | Up 1 point |
A one-point movement is smaller than the survey's margin of error, so these platforms are steady rather than shifting. Facebook remains a large audience for a local business. Google's Display campaign help page says the Display Network reaches people across websites and apps, including Google-owned properties such as YouTube and Gmail, so video inventory exists beyond Meta as well.
How Should a 15-Second Local Video Ad Be Structured?
No source reviewed sets an ideal length, so treat 15 seconds as a starting test rather than a rule. The structure below is a suggestion, not a platform requirement.
- Open within the first three seconds with who the ad is for and where, such as "Dog groomer in your town."
- Show one proof point you can film, such as a before-and-after, a menu item or a class in progress.
- Name one action in the final seconds, such as "Tap to book," and make the button match it.
- Put the main message on screen as text, so the ad still works with the sound off.
- Film on a phone in natural light, using real staff and your actual space.
Once organic posts go live alongside the ad, a scheduling tool keeps them consistent. A comparison of two options is in
Hootsuite vs Sprout Social for small local businesses.
Meta's specification pages for video ads were not among the sources checked for this guide, so this article does not state pixel dimensions, file limits or safe zones. Check the preview in Ads Manager before launch, and export a vertical and a square version if the placements you plan to use require them. Confirm current requirements in the platform's own help pages, not in third-party summaries.
How Do You Test Several Versions Without Skewing Results?
Meta's help page on A/B testing says the test shows each version to a segment of the audience so that nobody sees both. It recommends using the same budget for both versions, and it advises against informal testing, such as switching ad sets on and off by hand, because that can overlap audiences and make results unreliable.
Record one master video of 30 to 60 seconds, then cut three or four versions that differ only in the opening line, the first shot or the offer. Run them through Meta's A/B test tool rather than by hand. This changes one variable at a time without new filming.
What Budget Should a Local Test Get?
The reviewed sources do not include a current Meta cost-per-click benchmark. The most recent cost figure in the sources is a Google Ads average of $2.69 per search click across all industries, from a WordStream article first published in 2016. It is a Google figure, not a Meta one, so use it only as a rough reference. Your own cost per click after two weeks is the number that matters.
The example below uses assumptions for every input.
| Item | Value | Basis |
|---|
| Test budget for 30 days | $450 | Assumption |
| Cost per click | $1.50 | Assumption |
| Clicks | 300 | Budget divided by cost per click |
| Share of clicks that become leads | 5% | Assumption |
| Leads | 15 | Clicks multiplied by lead share |
| Share of leads who book | 25% | Assumption |
| Bookings | 3.75 (about 4) | Leads multiplied by booking share |
| Cost per booking | $120 | Budget divided by bookings |
| Value of a first booking | $70 | Assumption |
| Revenue from first bookings | $262.50 | Bookings multiplied by value |
At these assumptions, first bookings return about 58% of the spend. The test only pays if customers return, so estimate repeat visits before scaling. A local business that runs video on TV-style inventory can compare the same logic; see
CTV advertising for small businesses in Minneapolis.
FAQ
How long should a Facebook video ad be?
No source reviewed sets an ideal length. Fifteen seconds is a reasonable first test, and the structure above fits that length. Compare it with a 30-second version before deciding.
Should the video use captions?
Build the ad so the offer is readable on screen. A viewer who watches with the sound off should still see what the business does and what to do next. Test with captions on screen against a version without them.
How long should I run a test before judging it?
This guide suggests at least two to four weeks, a judgement based on typical test cycles rather than a platform rule. Judge on cost per result, not on views, and keep the budget fixed across versions.
Can a small budget work?
It can produce data, but slowly. A test needs enough budget to generate a handful of leads each week. If your budget cannot do that, the worked example above shows how few bookings a small test produces, so plan the budget before the video.
Sources
Want a local video test built and measured for you?
Book a free audit, or see
Google Ads services if search is part of the plan.
Free for local businesses
Want this applied to your business?
I'll review your Google presence, local SEO, and ad accounts — and send you a specific action plan within 48 hours. No pitch, no pressure.