What should a small business spend on Facebook ads, and what should it expect?
Start with a 30-day test of roughly $20 to $30 a day, which buys about 15 to 20 leads at the median cost per lead for home services in WordStream's 2026 benchmarks ($42.95). Expect leads from local people who weren't searching yet, not instant emergency calls, and judge the test on cost per lead and booked jobs, not likes.
The rest of this post explains where that budget comes from, how to set the campaign up, and how to read the results.
What Facebook ads are good at (and what they're not)
Facebook and Instagram ads create demand. They put your business in front of local people before they need you, so you're the name they remember when they do. Google search ads catch demand that already exists: someone typed "AC repair near me" and you show up.
That difference decides what Facebook ads are worth to you:
| Good fit | Weaker fit |
|---|---|
| A visual offer: food, remodels, salons, fitness, retail | Emergencies: burst pipes, locksmiths, towing |
| A seasonal offer or promotion | Services people only think about when something breaks |
| Staying in front of people for weeks or months | Catching someone who needs help in the next hour |
| Bringing back website visitors (retargeting) | Products nobody can picture from one photo |
Most local businesses need both kinds. For how Facebook fits alongside search, email and the rest, see our guide to the channels a small business can use, and what each costs.
What do Facebook ads cost?
WordStream's 2026 Facebook ads benchmarks, based on 1,892 US campaigns run between April 2025 and June 2026, put the medians at:
| Campaign type | Click-through rate | Cost per click | Cost per lead |
|---|---|---|---|
| Lead campaigns, all industries | 2.70% | $1.80 | $27.39 |
| Traffic campaigns, all industries | 1.93% | $0.60 | n/a |
| Lead campaigns, home and home improvement | $42.95 | ||
| Lead campaigns, dentists and dental services | $61.56 |
These are medians, so half of advertisers paid more and half paid less. Your costs depend on your area, your offer and, more than anything, your ad.
Boosted posts vs Ads Manager
The Boost button promotes a post already on your Page. It's quick, but it gives you simple goals and limited control.
Ads Manager is where real campaigns are built. It gives you:
- Lead forms that collect a name and phone number without leaving Facebook.
- Conversion tracking, so you can see calls and form fills from your website.
- Detailed location targeting by ZIP code, town or radius.
- Several ad versions at once, so you can see which message works.
Boosting is fine to give a good post more reach. For anything meant to bring in leads or sales, use Ads Manager.
How to set a first test budget
Work backward from leads, not from a round number:
- Pick your industry's typical cost per lead. Use the table above as a starting point, for example $42.95 for home services (WordStream, 2026).
- Decide how many leads you need to judge the test. Our working rule is at least 15 to 20 in 30 days. With fewer, one good or bad week can swing the result.
- Multiply. 15 to 20 leads × $42.95 is about $645 to $860 for the month, or roughly $20 to $30 a day over 30 days.
If that's more than you can spend, run the test longer rather than spreading a smaller budget thin. A test that can only buy five leads won't tell you anything.
One more reason not to go too small: Meta's documentation says an ad set needs about 50 results in seven days to finish its "learning phase", when the system works out who responds (Jon Loomer on the learning phase). Most small local budgets won't hit that, which is fine, but it's why you should avoid changing the campaign every few days.
How to set it up so it works
Target the area you actually serve. Use ZIP codes, towns or a radius. Choose people who live in or were recently in that area, not people "interested in" it, or you'll pay for clicks from out of town. We explain how radius targeting really works.
Keep the audience broad and let the ad do the targeting. Narrow interest settings shrink the pool Meta can find buyers in. A clear ad aimed at a specific customer ("Dacula homeowners: is your furnace ready?") finds the right people on its own.
Run three to five ads at once. Try different angles: the problem, the offer, a customer review, a short video of you or your team. Pause the weakest after two weeks.
Make the next step easy. A lead form with three questions, or a landing page with your phone number at the top. Then call back quickly, while the person still remembers filling in the form.
Install tracking before you launch. The Meta pixel on your website, plus a way to record which calls and forms came from ads.
What to expect after 30 days
Judge the test on these, in this order:
- Booked jobs or sales from the ads. The number that pays.
- Cost per lead. Compare it to the benchmark for your industry and to what a customer is worth to you.
- Lead quality. Did the leads answer the phone? Were they in your area? Did they want what you sell?
Ignore likes, reach and comments when you decide. They're useful for the person running the ads, but they don't pay the bills.
After 30 days you'll usually be in one of three places:
- Cost per lead near or below the benchmark and leads booking: keep it, raise the budget slowly, add new ads every few weeks so they don't wear out. See why your ad creative stops working.
- Leads coming in but not booking: the problem is usually the follow-up or the offer, not Facebook.
- Few leads at a high cost: change the ad and the offer before you give up on the channel.
Where Deep Thought fits
Deep Thought runs Facebook and Instagram ads for small businesses alongside search, streaming TV and the rest, all in one place and at a fraction of what an agency charges. Our team builds the campaigns, refreshes the ads and reports on leads and booked jobs in plain English. Want a starting plan? Our free digital media plan builder suggests a channel mix and budget split in a few minutes.
Sources
- WordStream, Facebook Ads Benchmarks 2026: 1,892 US campaigns, April 2025 – June 2026, medians.
- Jon Loomer, Is the Learning Phase 10 or 50 Events?: Meta's documented learning phase threshold.
Are Facebook ads worth it for a small business?
Usually, when your offer is visual or easy to explain and your customers are local. Facebook and Instagram ads are good at reaching people before they search and keeping you in mind. They are weaker for urgent needs like a burst pipe, where Google search ads usually win. Test for 30 days and judge on cost per lead and booked jobs.
How much should a small business spend on Facebook ads?
Enough to buy at least 15 to 20 leads in a 30-day test, so you can judge the results. WordStream's 2026 benchmarks put the median cost per lead at $27.39 across industries and $42.95 for home and home improvement, so a home services test of $20 to $30 a day for 30 days is a sensible start.
What is the difference between boosting a post and running an ad?
Boosting promotes a post that's already on your Page, with simple targeting and goals. Ads Manager gives you full control: lead forms, conversion tracking, detailed location targeting, several ad versions and placements. Boosting is fine for awareness of a post; use Ads Manager for leads and sales.
How long does it take for Facebook ads to work?
Give a campaign at least two to four weeks before judging it. Meta's system spends its first days learning who responds, and its documentation says an ad set needs about 50 results in seven days to finish that learning phase. Small budgets learn more slowly, so avoid changing the campaign every few days.
Should I target a radius on Facebook ads?
Yes, for a local business. Target the towns, ZIP codes or radius you actually serve, and choose people who live in or were recently in that area rather than people only interested in it. Leave the age and interest settings broad at first and let the ad's message do the targeting.