Facebook Ads and Your Website: Why One Without the Other Wastes Money
By Ricky Browning · Browning PC, Valdosta, GA
A business owner told me last year he'd spent about $600 on Facebook ads and gotten "nothing out of it." He wanted to know whether Facebook ads work. That's the wrong question, and it took me a while to figure out why. The ads weren't really the problem, and neither was the $600. The problem was that the money had nowhere to land.
Worth knowing: some of the AI-search statistics in this space were revised in early 2026 by the researchers who published them — I covered that reversal in this month's tech news roundup.
Here's the honest version up front: Facebook ads and a website aren't two separate purchases. They're two halves of one machine. The ad rents attention. The website turns that attention into a phone call — and it's the half you actually own. Buy one without the other and you're paying for a funnel with the bottom cut out.
Where I'm coming from, so you can weigh this properly: I build websites. That's a service I sell. I do not sell Facebook ad management, and nothing in this post is trying to get you to hire me for that. So read the website parts knowing I have a dog in that fight, and read the ads parts as what they are — a neighbor telling you what the numbers actually say before you spend money.
The handoff is where the money leaks
Most people picture it as a sequence: the ad does its job, then the website does its job. That's not how the platforms price it.
Google states plainly that Ad Rank is built from six factors, and one of them is the quality of your ads and landing page — and that "higher quality ads can often lead to lower CPCs." Landing page experience is also one of the three components of Quality Score. Read that again, because it's the whole argument in one line: a weak destination raises what you pay per click, before a single person decides whether to call you. You're not just losing conversions at the end. You're paying a worse price at the start.
It gets blunter. Google's Destination Requirements policy disapproves ads outright for "Destination not working," "Destination not crawlable," and "Destination not accessible" — including when the page simply returns an error to Google's crawler. A broken destination doesn't convert badly. It stops the campaign from running at all.
I'm citing Google here rather than Meta on purpose: Google documents this mechanism publicly and Meta doesn't publish an equivalent, so I'm not going to pretend I know Meta's auction works the same way. But the underlying logic — that the platform judges where you're sending people — is not a Google quirk. It's how mature ad auctions work.
What Facebook ads actually cost right now
Real numbers, with their warts attached.
LocaliQ's Facebook benchmarks, last updated October 24, 2025, put the all-industry averages for lead campaigns at $1.92 per click and $27.66 per lead, with a 2.59% click-through rate and a 7.72% conversion rate. Two things you should know before you lean on those: LocaliQ sells ad management, and it publishes no methodology for the Facebook figures — no sample size, no date range. Treat them as an order of magnitude, not a target.
There's a bigger catch for readers like you. That benchmark has no category for IT services, computer repair, or business services. There is no published number for a business like mine. The nearest rows are other trades: Home & Home Improvement at $41.26 a lead, Personal Services at $30.57. I'll use those as rough analogues, but substituting them is my judgment call, not something the report says.
And the spread across industries is enormous — dentists run $76.71 a lead while restaurants come in near $3. When the range is that wide, a single "average cost per lead" tells you almost nothing about your business.
One trend is solid because it comes straight from the source: in Meta's own Q2 2026 results, released July 29, 2026, the average price per ad rose 12% year over year while ad impressions rose 14%. That's a blended global figure across every advertiser and country, so it doesn't tell you your local price — but the direction is unambiguous. The same budget buys less than it did a year ago.
The math nobody shows you before you spend
This is the part I wish someone had explained to that $600 guy, because it explains his result completely.
Meta's guidance is that an ad set needs roughly 50 of its optimization events inside a rolling seven-day window before delivery stabilizes and it exits the "learning phase." Not 50 clicks — 50 of the specific event you're optimizing for. Fall short and Ads Manager flags the ad set "Learning Limited."
Now put the benchmark next to that. At roughly $28 a lead:
| Monthly budget | Leads per week | Share of the ~50/week guidance |
|---|---|---|
| $300 | ~2.5 | ~5% |
| $600 | ~5 | ~10% |
| $900 | ~7.5 | ~15% |
The exact cost per lead barely matters here. Even if your leads came in at half the benchmark, you'd still be at a fraction of the threshold. Meta has a further rule that bites locally, too: for any cost-per-result goal, your daily budget needs to be at least five times that target. Chase a $50 lead and Meta wants $250 a day before it will run the way it's designed to.
Then there's the part that's pure arithmetic and gets ignored entirely. Say a small campaign produces about 11 leads a month. Basic counting statistics — the square root of 11 is about 3.3 — mean a typical month lands within roughly 30% of your true average purely by chance. Compare one month to the next and swings of 40% are completely ordinary with nothing whatsoever having changed.
So when someone spends $600 and gets nothing, the honest diagnosis usually isn't "Facebook ads don't work." It's that the budget never generated enough signal for the system — or the owner — to learn anything from.
Speed is the cheapest thing you can fix
Most page-speed statistics you'll see are correlations dressed up as cause. Here's the exception, and it's why I lead with it.
In 2021, Vodafone Italy ran an actual A/B test: two versions of a page, visually and functionally identical, one optimized for loading speed. The faster version — a 31% improvement in Largest Contentful Paint — produced 8% more sales, an 11% better cart-to-visit rate, and a 15% better lead-to-visit rate. One company, one market, and they published it themselves. But it's a controlled test rather than a survey, which puts it in a different class from almost everything else written on this subject.
Google's current standard is a Largest Contentful Paint of 2.5 seconds or less, with anything over 4 seconds rated poor — measured at the 75th percentile, meaning three out of four visits need to hit it, not just your average one. That's the real benchmark, and it quietly replaces the folk "three-second rule," which has no official standing anywhere.
Worth knowing how common the problem is: as of July 2025, only 48% of mobile origins passed all three of Google's Core Web Vitals, versus 56% on desktop. Loading speed is the binding constraint — 62% of mobile origins had good LCP, compared with 77% for interactivity and 81% for visual stability.
Your ad traffic is overwhelmingly mobile. If your site is in the slow half, you're paying full price for clicks that bounce before the page paints.
A Facebook Page is not a website, and Meta says so in writing
Plenty of local businesses run a Page instead of a site. It's free, it's familiar, and it feels like enough. Here's what you're actually agreeing to.
Read Meta's Terms of Service. Meta "may suspend or permanently disable your access to Meta Company Products." It states that "we are not liable to you for any lost profits, revenues, information, or data." And its total liability to you "will not exceed the greater of $100 or the amount you have paid us in the past twelve months." That's the deal. If your entire customer-facing presence lives there and it goes away, the contractual remedy is a hundred dollars.
This isn't hypothetical. Meta's own newsroom reported taking down more than 100 million fake Pages in 2024, plus over 23 million impersonation profiles, and around 10 million more impersonating profiles in the first half of 2025. Enforcement at that scale has to be automated — and any automated system making hundreds of millions of decisions produces false positives.
Some of those false positives are real businesses. An Australian gym owner, Haylee Toohey, was locked out for nearly five weeks, losing two to three leads a week. Starting over wasn't an option: "I created a new account," she told Yahoo Finance Australia in August 2025, "but anything with my business name was immediately shut down." And paying for support is no guarantee — one locked-out Meta Verified subscriber pointed out the obvious flaw: "The enhanced support was inaccessible because I could not log into my account." By March 2026, CBS News Philadelphia reported nearly 60,000 people had signed a petition over suspensions, with users nationwide describing the same experience.
There's a quieter problem too, and you can verify this one yourself in about ten seconds. Pull up facebook.com/robots.txt. It names 33 crawlers, restricts each of them, then finishes with User-agent: * / Disallow: / — everything else refused. Training crawlers like GPTBot and ClaudeBot are on the named list. But the fetchers that go read a page live to answer a question someone just typed are not.
That matters more every month. BrightLocal's 2026 Local Consumer Review Survey of 1,002 US consumers found the share using AI tools like ChatGPT to find local businesses jumped from 6% to 45% in a year, while Google's share fell from 83% to 71%. It's a survey of what people say they do, so hold it loosely — but if a growing slice of people are asking an AI for a recommendation, a Facebook Page is a locked door and your website is not.
And Google is blunt about the ceiling. In its own documentation on using a Business Profile as an ad landing page, it says a Business Profile "isn't a replacement for your business's website," and that it can directly attribute only two actions to an ad: a phone call, or a lead-form submission. Everything past that is simply unmeasurable.
If it were my money, here's the order
Not because ads are bad — because sequence determines whether the ad money does anything.
Google Business Profile and reviews, first
Whitespark's 2026 study — 47 local SEO practitioners scoring 187 factors — ranks the top Local Pack factors as your primary GBP category, proximity to the searcher, keywords in your business title, a physical address in the city being searched, and being open when someone searches. That's expert opinion rather than a measured experiment, but it's free to act on. And reviews now carry real weight: BrightLocal's 2026 survey found 31% of consumers say they'll only use a business rated 4.5 stars or higher — nearly double the 17% who said so a year earlier — 47% won't consider a business with fewer than 20 reviews, and 74% want reviews from the last three months.
A site that loads fast and asks for the call
This is the destination every other channel depends on — ads, search, the AI answers, the business card you hand someone. Fast, clear about what you do and where, and obvious about how to reach you. It's also the only piece in this whole list that you own outright.
Then ads — with a budget that can actually teach you something
Once there's somewhere good to send people and a way to tell whether it worked. If the budget you have in mind lands in that $300-a-month row above, my honest advice is to put it into the first two steps this year and come back to ads later. That's not me talking you out of spending money. It's me telling you what the arithmetic says.
Statistics you'll see in every other article about this — and shouldn't trust
I had these fact-checked hard before writing, and a startling number of the "facts" that show up in posts on exactly this topic fell apart. Since you'll run into them, here's what's wrong with them.
| The claim you'll see | What's actually true |
|---|---|
| "75% of people judge a company's credibility by its website design" — usually credited to Stanford | That figure appears in no Stanford publication. The real study (Fogg et al., Stanford, October 2002) had 2,684 people write 2,440 comments; "design look" came up in 46.1% of them. It's also 24 years old and predates mobile entirely. |
| "Businesses make $4 for every $1 spent on Facebook ads" | No source. Meta publishes no such figure. For scale, the analytics vendor Triple Whale found a median return of 1.86x across nearly 35,000 e-commerce brands in 2025 — and that's e-commerce, not a local service business. |
| "70% of small businesses that fail lack a website" | Fabricated. No study connects the two. It appears to splice government business-closure data onto unrelated website-adoption surveys. |
| "27% of small businesses don't have a website" | One publisher family reported 37% in 2020, 28% in 2021, 27% in 2022, and 17% in 2025 — a twenty-point swing across small opt-in panels. Adoption doesn't move that fast. Anyone quoting one of those as settled fact is quoting one panel on one day. |
| "53% of mobile users abandon a site after 3 seconds" | This one's real but mangled. Google/DoubleClick, September 2016, from data collected that March: it's 53% of visits, not users, across ~3,700 opt-in sites. It's a decade old. It also gets restated in a 2017 Akamai report, so citing "both Google and Akamai" makes one finding look like two. |
I'm including this section because it's the most useful thing here. If an agency pitches you using any of the above, you've learned something important about how carefully they check their work before spending your money.
The short version
- The ad and the destination are priced together. Google says landing page quality is one of six Ad Rank factors and that better quality often means lower cost per click. A bad destination costs you twice.
- Small budgets fail for a structural reason, not a creative one. Meta wants roughly 50 events a week to optimize; $300 a month buys about 5% of that.
- At low volume, month-to-month swings of 30–40% are chance. Don't rebuild a campaign in response to noise.
- Speed is the cheapest lever you have. Target LCP under 2.5 seconds at the 75th percentile. Only 48% of mobile sites pass all three Core Web Vitals.
- A Page is rented, and the lease is harsh. Meta's liability cap is $100, its enforcement is automated at enormous scale, and Pages are invisible to the live AI crawlers that 45% of consumers now use to find local businesses.
- Order matters: Google Business Profile and reviews, then the website, then ads.
🌐 Need the destination half sorted out?
That's the part I actually do. I build fast, straightforward sites for small businesses around Valdosta and South Georgia — the kind that load quickly on a phone and make it obvious how to reach you, so that whatever you spend on getting attention isn't wasted when it arrives. Have a site already? Start with the seven signs it's costing you customers, or run it through our free Website Growth Scanner. Then see what a build involves, or just call and ask.
📞 229-561-1674 ·
📅 Book an appointment
Frequently Asked Questions
How much should a small business spend on Facebook ads per month?
There's no sourced answer, and anyone who gives you a confident number is guessing. What's documented is the constraint: Meta wants roughly 50 optimization events per ad set per week, and at the published benchmark of about $28 a lead, that works out to somewhere near $6,000 a month on a single ad set. Most local businesses aren't spending that. That doesn't mean don't advertise — it means expect a small budget to produce noisy results you can't easily learn from.
Can I just use a Facebook Page instead of a website?
You can, and you're accepting three specific costs. Meta's Terms cap its total liability to you at $100 and disclaim any responsibility for lost revenue. Facebook's robots.txt blocks the live AI crawlers that answer people's questions, so your Page is invisible there while 45% of consumers now use AI to find local businesses. And Google's own documentation says a Business Profile "isn't a replacement for your business's website." A Page is a good front door. It's a bad foundation.
Does website speed really affect my ad costs?
On Google, yes, and it's documented: landing page quality is one of six Ad Rank factors and one of three Quality Score components, and Google states higher quality can lead to lower costs per click. Ads can also be disapproved outright if the destination errors out. Meta doesn't publish an equivalent mechanism, so I won't claim it works identically there — but on the conversion side, Vodafone Italy's controlled A/B test found a 31% speed improvement produced 8% more sales.
Facebook ads or Google — which is better for a local service business?
For most local service businesses, I'd get the free search fundamentals right before paying for either. Someone searching "computer repair near me" already wants the thing; someone scrolling Facebook doesn't. Google Business Profile and reviews cost nothing but effort and are where local discovery still concentrates — 71% of consumers per BrightLocal's 2026 survey. Paid social is better at creating demand than capturing it, which makes it the second move, not the first.
One last thing
If you take nothing else from this: the question isn't "do Facebook ads work." It's "what happens to someone after they tap the ad?" Answer that honestly and you'll usually find the cheapest fix isn't in the ad account at all.
And if you're around Valdosta, Hahira, or anywhere in South Georgia and want a straight opinion on whether ads make sense for your situation — just ask me. I'll tell you if I think the answer is no. I'd rather do that than watch another neighbor spend $600 finding out.