How Much Do Facebook Ads Cost in 2026? Real Numbers, Not Averages

Search "how much do Facebook ads cost" and you get the same sentence on every page: the average CPC is $0.97 and the average CPM is around $14. Those numbers are real. They are also useless to you.

Here is why. That average blends a mobile game company buying installs for pennies with a personal injury firm fighting for homeowners in Dallas, and a t-shirt store retargeting cart abandoners with a roofer looking for houses that have a 20-year-old roof. Averaged together, they describe no one. It is like being told the average vehicle costs $48,000 when you came in to buy a work truck.

This page does the opposite. Real ranges by trade, with the job value next to the lead cost so the number means something — and a straight answer to the question you came here with: what do I have to spend to make the phone ring?

Why the "average CPC" is the wrong number

Your cost is not set by an industry average. It is set by an auction that runs every time someone near you opens Facebook or Instagram — priced by how many advertisers want that person's attention, and how good Meta thinks your ad is at holding it.

That means two plumbers running the same offer on the same budget in the same week can see costs differ by three times. Not because one is doing something wrong — one is bidding against a dozen home service companies in Phoenix, the other is the only advertiser in a county of 40,000 people.

So drop the average. There are only four numbers that matter, and they stack on top of each other:

• CPM — what you pay for 1,000 people to see your ad. For local US audiences in 2026, budget $8 to $30, with tight metros and competitive verticals running $35-50.
• CPC — what you pay when someone taps. Local small business: $0.70 to $3.50. This is a derived number, not a dial you set.
• Cost per lead (CPL) — what you pay for a name and a phone number. This is where most owners should stop looking at anything else.
• Cost per booked job — what you pay for money in the bank. This is the only number that decides whether ads are working.

Watch how they chain together. Say you pay a $25 CPM and 1% of viewers tap: 1,000 impressions costs $25 and produces 10 clicks, so your CPC is $2.50. If 10% of those clickers complete your form, you got 1 lead for $25. Now change nothing but the creative and the offer — CTR drops to 0.5% and form completion drops to 4%. Same $25 CPM, same audience, same budget. Your CPC is now $5.00 and your CPL is $125. Five times more expensive, and Meta charged you exactly the same rate to be seen.

That is the whole game. You have almost no control over CPM. You have enormous control over everything downstream of it.

What a lead actually costs, by trade

Honest ranges for US local businesses running lead-form campaigns. The low end is a strong offer in an uncompetitive market; the high end is a generic offer in a metro. The "minimum viable budget" column matters most — it is what it takes to buy roughly 25-30 leads in 30 days, the smallest sample you can actually read.

Two things jump out of that table. First, the spread inside a single trade is enormous — a roofer can pay $45 or $120 for the same thing. Second, the trades with the most expensive leads are the ones where advertising works best. That is not a coincidence, and it is the most important idea on this page. For the full cost-and-offer playbook on your own trade, see our guides for roofers, HVAC companies, plumbers, contractors, electricians, landscapers, dentists, gyms, restaurants, cleaning companies and realtors. If you want the spend framed as a monthly plan rather than a per-lead price, how to budget for Facebook ads lays it out.

A $150 lead can be cheaper than a $15 lead

Owners fixate on lead price because it is the number Meta puts in front of them. It is the wrong thing to optimize. Here is the arithmetic, twice.

Example A: a roofer in Tulsa paying $150 a lead

He spends $1,800/month. At a $150 CPL that buys 12 leads. He reaches 8 of them and books inspections. Three sign. So:

• Cost per booked job: $1,800 / 3 = $600
• Average job: $11,000, at 35% gross margin = $3,850 gross profit
• Monthly gross profit from ads: 3 × $3,850 = $11,550
• Ad spend per dollar of gross profit: $1,800 / $11,550 = 16 cents

He pays $150 per lead and it is one of the best deals in his business. The first signed roof pays for six months of advertising.

Example B: a barbershop paying $15 a lead

She spends $600/month. At a $15 CPL that buys 40 leads. Roughly 45% actually book and show up — 18 new clients.

• Cost per new client: $600 / 18 = $33
• First haircut: $35, about $20 in margin after chair time
• First-visit revenue: 18 × $20 = $360 on $600 spent. Month one loses $240.
• Say 4 of the 18 become regulars: 8 visits/year × $20 = $160 each = $640 over twelve months
• Year-one gross profit: $360 + $640 = $1,000. Ad spend per profit dollar: 60 cents.

Her leads cost one tenth of his. Her advertising is roughly four times less efficient, and it takes her the better part of a year to break even on a month of spend that the roofer recovers in week two.

The rule that falls out of this: never judge a lead price without the job value next to it. If your average job is worth $8,000, a $150 lead is a bargain. If your average ticket is $35, a $15 lead can quietly bleed you. That is the same math behind the question of whether $10 a day can work — for a roof, absolutely; for a haircut, rarely.

What actually moves your cost up or down

Five things, ranked by how much they matter for a local business:

1. Offer strength. The biggest lever and the cheapest to pull. "Free roof inspection, photos of every problem we find, no obligation" beats "Get a quote" by a mile. Owners routinely cut CPL 30-50% by changing nothing but what they promise.
2. Creative quality. A real photo of your crew, your truck, your before-and-after beats a stock image almost every time. Stock photography is the single most common reason a good offer produces a bad CTR — and remember, halving CTR doubles your CPC.
3. Audience size. Below roughly 50,000 people, Meta runs out of room and shows your ad to the same faces repeatedly. Frequency climbs, performance dies, CPM inflates. If your ZIP radius is too tight to hit that number, widen the radius rather than narrowing the interests.
4. Competition in your ZIP. A cleaning company in suburban Austin might pay a $28 CPM. The same company in a rural county 90 minutes away might pay $11. Same ad, same offer. You cannot fix this — you can only account for it when you set the budget.
5. Season. CPMs rise 25-40% from Black Friday through Christmas as retailers flood the auction, then fall in January. HVAC in July and landscaping in April are competing against every other HVAC and landscaping company in town. Buying demand 4-6 weeks early is a real discount.

Notice what is not on that list: the bid strategy, the placement checkboxes, and the twenty other settings in Ads Manager that people argue about. They are rounding errors next to the offer.

When the honest answer is "don't spend it here"

Some money should not go to Facebook at all, and this is the part most cost articles skip.

If your work is emergency work — burst pipe, no AC in August, tooth pain — Facebook is the wrong auction. That customer is typing into Google right now and calling whoever answers. Paying $30 to show them a feed ad they will scroll past is a donation. Put that budget in your Google Business Profile and Local Services Ads. Our Facebook vs Google comparison lays out which demand belongs where.

If your average ticket is under about $75 and customers don't come back, the math almost never closes. See the barbershop above; a one-visit-only business at a $35 ticket has no room to pay for acquisition.

And if you have fewer than 30 days of budget, don't start. Meta needs roughly 50 conversions per week per ad set to optimize properly. Most local businesses never reach that, which is survivable — but you still need enough volume to tell signal from noise. Two weeks at $10/day tells you nothing, and you will conclude ads don't work when all you learned is that $140 buys three leads.

So what do you need to spend to make the phone ring?

Working backwards from the table, here is the plain answer for a 30-day test that gives you a real verdict:

• Roofing: $1,500-2,200/month. Expect 15-30 leads, 3-6 booked jobs.
• HVAC replacement: $1,200-1,800/month. Expect 15-30 leads.
• Plumbing (planned): $750-1,200/month. Expect 20-35 leads.
• Remodeling contractor: $1,800-2,700/month, and be patient — these close on a 30-60 day cycle.
• Realtor: $750-1,200/month. One closing pays for the year.
• Dental (implants/ortho): $1,200-2,100/month. Two cases covers it.
• Gym: $450-750/month. Expect 25-60 leads; judge on 90-day retention, not signups.
• Landscaping / cleaning: $450-900/month. Recurring contracts make this the easiest math on the list.
• Restaurant: $300-600/month, and fix your Google listing and your photos first.

If those numbers feel high, that is the honest reality of the channel — better to know now than after burning $200 on a boosted post. Our guide to setting a Facebook ads budget works the same logic from the other direction, starting from what a job is worth to you.

The cost you forgot: running the thing

Everything above is media spend — money that goes to Meta. It is not what the ads cost you. And that spend only turns into booked jobs if the follow-up holds up: getting the lead out of Facebook and onto the phone fast matters as much as the number, and if your cost per lead sits stuck at the high end of the ranges above, why your Facebook ads aren't working is usually the reason.

Do it yourself and you pay in hours. Ads Manager is not hard so much as cluttered, and the first campaign realistically costs a weekend plus a few hundred dollars in beginner mistakes. Fair price, if you have the weekend.

Hire an agency and US pricing lands around $1,000-3,000/month retainer or 10-20% of spend, usually with a $2,000/month spend minimum. The math is brutal at small budgets: a $1,500 retainer on top of $1,500 in spend means half your money never reaches a customer. Agencies genuinely earn their fee above roughly $3,000/month in spend — we break down exactly where the line sits in our piece on what Facebook ads agencies charge.

The third option is a tool that does the work. Leadria uses AI to build and publish your Facebook and Instagram ads: you describe your business in a sentence — "roofing contractor in Tulsa, full replacements and storm damage" — and the AI writes the copy, generates the visual, sets the Meta targeting, and publishes the campaign. Leads come back with a phone number attached. There is a 7-day free trial and no credit card required, so you can see what it produces before you decide.

Whichever route you take, hold onto the one number that matters: cost per booked job, measured against what that job is worth. Not CPC. Not likes. If you want the full picture of how the channel fits a local business, start with our complete Facebook ads guide for small business — and then go find out what a lead costs in your ZIP code, because that is the only benchmark that was ever going to matter.