Nobody on Facebook is looking for you.
They're watching dog videos. They're in line at Dutch Bros. They're half-listening to a meeting. Nobody has ever opened the app hoping to see an ad about gutter cleaning.
That's the context every ad you run has to survive. And it's why most local ads fail before anyone reads a word of them — not because the targeting was wrong, but because the ad was written for someone who was already interested.
If you take one thing from this: your ad's job is not to sell your service. It's to sell the click.
Ad Sells the Click. Landing Page Sells the Product.
This is where most local businesses lose money, and it's not obvious.
They try to make the ad do everything — grab attention, explain the service, build trust, list the pricing, overcome objections, and close. All in a square image somebody sees for a second and a half.
That doesn't make the ad stronger. It makes it heavier. And heavy ads get scrolled past.
The ad has one job: spark enough curiosity that clicking feels worth it. That's all. Enough interest to get a "huh, what's this?"
Everything else happens after the click. That's the landing page's job — explain, prove, offer, convert. It has the reader's attention, it can take thirty seconds instead of one.
Keep those roles separate and everything downstream gets easier. Writing ads gets faster because you're only trying to earn one small action. Testing gets cleaner because you're measuring one thing. Scaling gets simpler because you know which half is broken when results drop.
Try to win the whole sale in the ad and you'll usually lose the click too.
Write Like You'd Say It Out Loud
Your customer skims. They don't read, and they definitely don't read industry language.
Compare these:
"Professional HVAC tune-ups available now in Vancouver."
"AC blowing warm air? We've got next-day appointments in Vancouver."
The second one works because it starts where the customer already is — with the problem they're actually having — and because it sounds like a person said it.
A few rules that hold up:
- Lead with the pain, then the fix. Not your service name.
- Use their words. If customers say "gunked-up gutters," write that. Don't upgrade it to "debris removal."
- Contractions. "You're," not "you are." You're texting, not filing a permit.
- Two or three short sentences. That's the whole budget.
- Name the place. "Serving SE Portland near Hawthorne" does more for trust than any adjective.
The test: would you say this sentence to your cousin over coffee? If not, don't run it.
Match the Format to the Trust Problem
Not every business needs video. The format should do whatever builds trust fastest for your service:
- Chiropractic — a short clip of an actual adjustment, or a patient describing what changed. Movement and sound do the convincing.
- Landscaping — before-and-after carousel. The gap between the two photos is the entire pitch.
- Cleaning — a time-lapse, or honestly just one photo of a spotless kitchen.
- Detailing — before-and-after, every time.
If your work looks dramatically different when it's finished, show that. If your credibility comes from a person, show the person. Don't produce a video because someone told you video performs well.
Stop Boosting Posts
Boosting is the button Facebook puts in front of you because it's easy, not because it works.
You get reach. You don't get zip-code targeting, retargeting, proper objectives, or the reporting that tells you what actually happened.
Ads Manager is more to learn, and it's where the tools live:
- Target specific zip codes instead of a whole state
- Show ads to people who already visited your site
- Run follow-up ads to people who watched but didn't act
- See cost per result, not cost per impression
That's the line between "we tried Facebook ads" and a business getting leads every week.
Target Like a Local Business, Not a National One
If you serve Vancouver, targeting "Washington" is fishing the whole ocean for one salmon.
- 5–10 mile radius around your actual service area. If you won't drive there, don't advertise there.
- Put the city in the copy. "Vancouver homeowners" stops the scroll for people in Vancouver and costs you nothing.
- Stack interest with location where it makes sense — parents in Vancouver who follow home improvement, for instance.
This is the advantage you have over a national brand running one generic ad in forty markets. Use it.
Budget: Start Small, Stay Focused
The question everyone asks is how much to spend. For most local service businesses, $10–20 a day is enough to learn something without lighting money on fire.
The bigger mistake isn't the number — it's spreading it too thin. One service, one offer, one area. Not "all our cleaning services across Portland," but move-out cleaning in NE Portland with a real reason to book this week.
- Let it run 5–7 days before you change anything. Changing daily just resets the learning.
- Track cost per result, not clicks — cost per call, per form, per booking.
- Set a real target. Ten calls this week. Five bookings. If you don't name it, you can't tell whether it worked.
What matters isn't the spend. It's the ratio between spend and what came back.
Test One Thing at a Time
No campaign is right on the first try. The winners get found, not written.
Test in pairs — two headlines, or photo versus video, or two different calls to action. One variable at a time, or you'll never know which change did what.
The changes that move results are usually smaller than you'd think. Swapping an image of a worker on a ladder for a close-up of the finished result can shift performance meaningfully, because one shows your job and the other shows their outcome.
Treat it like an experiment, not a launch. Controlled variables, deliberate changes, everything tracked.
Where the Money Actually Leaks
Even a good ad dies if the click lands somewhere bad. The most common versions:
- The page takes ten seconds to load. They're gone before it renders.
- The ad promised an estimate; the link goes to your homepage. Now they have to hunt for what you just offered them.
- The form is painful on a phone. Which is where nearly all of this traffic is.
- No obvious next step. If you want calls, the number should be impossible to miss.
- Nobody follows up fast. This is the big one, and it wastes everything upstream of it.
Every one of those is cheaper to fix than a bigger ad budget, and fixing them makes the same ad perform better without spending another dollar.
Common Questions
How long before I change something?
Five to seven days minimum. The algorithm needs time to find your people, and daily tinkering resets that. If there's genuinely no traction after a week, change one variable — image, headline, or audience.
Should I run on Instagram too?
Usually yes, and it's a toggle inside Ads Manager. It tends to matter more for visual services — cleaning, organizing, landscaping, detailing — and for younger homeowners.
How do I know it's working?
Count bookings, calls, and messages, not clicks or likes. Then compare cost per result against what a customer is actually worth to you. A $60 lead is expensive for a $80 job and cheap for a $4,000 one.
Do I need a professional video?
No. A clear photo of real work with a plain headline often beats produced video, because it looks like it came from a real local business instead of an agency. Start simple.
What if I've already tried and it didn't work?
Then something specific broke, and it's worth finding out what. Usually it's the offer or the follow-up rather than the ads — we wrote about that here.
