You have a budget that is real but not big. Everyone tells you to run both. Nobody tells you that running both is exactly how small budgets get wasted.
This guide gives you one answer. Three questions, a breakdown by business type with real 2026 cost data, and an honest note on when adding the second platform actually makes sense.
The short answer
If people are already searching for what you sell, start with Google Ads. If they are not, if your product is something people want once they see it rather than something they go looking for, start with Facebook Ads.
Google captures demand that already exists. Facebook creates demand that does not. Almost every small business fits clearly into one of those two situations, and the rest of this article helps you work out which.
Cheap clicks are not cheap leads
Facebook clicks cost less than Google clicks. That is true, it is why most small businesses try Facebook first, and it is the most expensive misunderstanding in small business advertising.
What matters is not what a click costs. It is what a customer costs.
Ruler Analytics tracked over 110 million sessions and 5 million conversions across 13 industries. Their finding: paid search converts at 5.4% on average, paid social at 2.11%. In professional services the gap is wider still, 6.7% against 1.1%.
So if your Facebook clicks cost half as much but convert at a third of the rate, you are paying more per customer, not less. The dashboard looks better. The bank balance does not.
| [HIGHLIGHT BOX] THE ONE METRIC THAT MATTERSJudge platforms on cost per customer, never on cost per click. Those two numbers often point in opposite directions and cost per click is the one both platforms put at the top of your dashboard. |
Why “just run both” wastes a small budget
Here is the part nobody explains.
Both platforms run on automated bidding, and automated bidding is a learning system. Learning systems need volume. Meta says it plainly: an ad set generally needs around 50 optimisation events a week to exit its learning phase and perform predictably. Google’s Smart Bidding has similar appetites.
Now run the numbers on a real small business budget. The median cost per lead across all industries is $66.69 (WordStream, 2026). At that rate, $1,500 a month buys you roughly 22 leads. That is five a week. You will never reach 50.
So the account gets set to optimise for something cheaper and more frequent page views, or form opens, or button clicks. The algorithm now has enough volume, and it gets very good at finding people who open forms. Your cost per “lead” drops. Your cost per actual customer does not move. Plenty of businesses read that as success for months.
Split the same $1,500 across two platforms and you have two campaigns with eleven leads each, neither learning anything. That is not diversifying. That is building two half-campaigns instead of one working one.
| THE RULE: Under roughly $3,000 a month, run one platform properly. One funded platform beats two starved ones — every time, and it is not close. |
Three questions that pick the platform for you
Answer these in order. Ten minutes total, and the first one requires you to look something up rather than guess.
1. Are people already searching for what you sell?
This is the question almost nobody checks, and it settles most cases on its own.
Open Google Keyword Planner, it is free with any Google Ads account and you do not need a live campaign. Choose “Discover new keywords”, enter the things you actually sell, and set the location to your real service area rather than the whole country. Then add up the monthly searches for your top five terms.
| Monthly searches | What it means | Points to |
|---|---|---|
| 300+ | Real demand exists. People are looking for this every week. | |
| 100–300 | Thin but workable. Expect a slower trickle. | Google, patiently |
| Under 100 | There is nothing to capture. Bidding harder will not create searches. |
If nobody is searching, Google Ads cannot help you. Not because the platform is weak, but because there is no demand there to capture. That is precisely the situation Facebook exists for.
2. Do they need it now, or want it later?
Urgency creates search. Nobody scrolls Facebook hoping to stumble across an emergency plumber while water comes through the ceiling, they pull out a phone and type.
| Situation | Examples | Points to |
|---|---|---|
| Needed today | Burst pipe, no heat, emergency dentist, locksmith, car breakdown | |
| Needed soon | Appliance repair, pest control, accountant, moving company | |
| Wanted eventually | Kitchen remodel, gym membership, new sofa, holiday booking |
3. Does it sell on sight?
Facebook is interruption advertising. It works when someone can want your product the moment they see it — which means it needs something worth seeing.
A bathroom before-and-after sells on sight. A dress sells on sight. A cleared drain does not. If your best photograph is of a clean pipe, Facebook has very little to work with.
What this means for your type of business
Cost figures below are median Google Ads benchmarks from WordStream’s 2026 report, covering 13,474 US search campaigns. The all-industry medians for comparison: $5.42 per click, 8.18% conversion rate, $66.69 per lead.
1. Home services and trades
Google, almost always. Median cost per click is $8.33 and cost per lead $90.92 expensive on paper, but conversion runs at 8.05% because the intent is real. Someone searching “emergency electrician” is not browsing. Save Facebook for planned work like remodels, where before-and-after photos do the selling.
2. Dental practices and clinics
Google. Dentists convert at 10.67%, well above the 8.18% average, at $72.97 per lead. People search for dentists by need and by location, and they book quickly. Facebook has a narrow role for cosmetic work veneers, whitening, aligners — where the result is visual and the decision is elective.
3. Salons, spas and beauty
Both work, and this is one of the few genuinely close calls. Google is efficient here $4.62 per click, 10.35% conversion, $39.25 per lead. But beauty is visual and social by nature, so Instagram and Facebook earn their place. If you are choosing one to start: Google for services people search for, Facebook for treatments they discover.
4. Restaurants and food
Facebook first, with one caveat. Google is remarkably cheap for restaurants $2.05 per click and $30.57 per lead — but most restaurant discovery happens on Maps and social, not paid search. Put your energy into your Google Business Profile and local social, and use paid search for catering, private events and delivery.
5. Ecommerce, apparel and online stores
Facebook, usually. Apparel and fashion convert at just 4.50% on Google with a $97.51 cost per lead roughly half the average conversion rate at nearly 1.5× the cost. That gap exists because clothing is discovered, not searched. Facebook and Instagram are built for exactly that. Google Shopping earns its place once people search for your brand by name.
6. Gyms, studios and memberships
Facebook first. Health and fitness converts at 6.94% on Google at $67.36 per lead, respectable, but people rarely search for a gym; they join one because something prompted them. Transformation content, class clips and local community targeting do more work than keywords. Use Google for “gym near me” and the January rush.
7. Law firms
Google, without hesitation. Legal is the most expensive category in the data, $9.87 per click and $131.63 per lead, but the case value justifies it and legal need is urgent and searched. Facebook is nearly useless for acquisition here; nobody scrolls their way into hiring an attorney.
8. B2B and business services
Google, then LinkedIn; not Facebook. Business services convert at 4.85% on Google at $93.69 per lead. Buying cycles are long and committees are involved, so expect to pay for a lead that takes months to close. Facebook targets people, not job titles, which is why it rarely works for B2B.
| IF YOU ARE A LICENSED TRADE, CHECK THIS FIRST: Plumbers, electricians, HVAC, roofers, locksmiths and garage door companies have a third option most people skip. Google’s Local Services Ads sit above regular search results, carry the Google Guaranteed badge, and charge per lead rather than per click. They need licence and insurance verification, which is exactly why fewer competitors bother. For many trades they beat both platforms on cost per booked job. |
What each platform costs to run properly
Underfunding either platform produces the same outcome: enough spend to lose money, not enough to learn anything.
1. Google Ads
The realistic minimum
$1,500 a month in most industries, more in legal or home services where clicks run $8–10. Give it six to eight weeks before judging. Payback is fast, often inside the first month.
What you need before you start
- A page for each service that matches what people searched for, not your homepage for everything
- Call tracking, so you know which clicks became phone calls
- Someone who answers the phone during business hours
Google Ads usually fails because of what happens after the click, not the click itself.
2. Facebook Ads
The realistic minimum
$1,000 a month, and eight to twelve weeks before you can judge it fairly. Payback is slower because you are creating demand rather than catching it.
What you need before you start
- Real photography or video; before-and-afters, your team, your actual work
- An offer worth stopping a scroll for
- A landing page and a follow-up process, because these leads are colder
Facebook usually fails because of what you brought to it, not what it did.
When running both finally makes sense
“Run both” is not wrong forever. It is wrong too early. Add the second platform when all three of these are true:
- Your first platform is consistently profitable and you know your cost per customer, not just per lead.
- You can fund the second without dropping the first below its working level realistically $3,000 to $4,000 a month total.
- You can handle more work. More leads into a business already at capacity means slower responses and worse reviews.
The sequence is always the same: win on one, prove the numbers, then fund the second out of profit.
Your first 90 days
| Weeks | If you chose Google | If you chose Facebook |
|---|---|---|
| 1–2 | Exact and phrase match only. Add negative keywords daily. Set up call tracking before you spend a dollar. | One audience, one offer, three creatives. Photography before copy, the image does the work. |
| 3–6 | Cut search terms that never convert. Build a landing page per service. | Let it run. Every significant edit restarts the learning phase. |
| 7–12 | Scale what works. Start measuring cost per customer, not cost per lead. | Test one new audience against your winner. Keep the offer fixed so you know what changed. |
One habit matters more than anything in that table: measure cost per customer. Cost per lead is the number both platforms show you, and it hides the difference between someone who filled in a form and someone who paid you.
The short version
- Judge platforms on cost per customer, never cost per click.
- Under $3,000 a month, run one platform properly instead of two badly.
- People already searching means Google. Nobody searching means Facebook.
- Spend ten minutes in Keyword Planner before you spend a dollar on ads.
- Licensed trade? Check Local Services Ads before either.
- Add the second platform out of profit, not out of a split budget.
| Still not sure which way your business points? We will run the three questions with you — real search volume for your area, a look at what you are currently paying per customer, and a straight recommendation on which platform to fund first. You keep the analysis either way. Get Your Free Ad Platform Assessment → /contact/ |
Frequently asked questions
Q1: Which is cheaper for small business, Google Ads or Facebook Ads?
A. Facebook is cheaper per click and often more expensive per customer. Paid search converts at around 5.4% on average against 2.11% for paid social. Half the click cost at a third of the conversion rate is not cheaper — it just looks cheaper on the dashboard.
Q2: Can I run Google Ads and Facebook Ads on a $1,000 monthly budget?
A. You can, but you should not. Both platforms need conversion volume to optimise — Meta documents roughly 50 optimisation events per ad set per week. At a $66.69 median cost per lead, $1,000 produces about 15 leads a month. Split in two, neither campaign learns anything.
Q3: How long before Google Ads starts producing leads?
A. Often within two to four weeks if there is real search demand in your area, though give it six to eight weeks before judging the data. Facebook usually takes eight to twelve weeks because it is building demand rather than catching it.
Q4: What if nobody searches for my product?
A. Then Google Ads cannot help you, and that is a demand problem rather than a platform problem. Check properly in Google Keyword Planner using your real service area. Under 100 monthly searches across your top five terms means Facebook is the better starting point.
Q5: Is Facebook Ads worth it for local businesses?
A. It depends on whether your service is visual and discretionary. Salons, gyms, restaurants and remodelling businesses do well because the result photographs well and the purchase is elective. Emergency and repair services do poorly, because nobody discovers a plumber while scrolling.
Q6: Should I stop Facebook Ads if they are not producing leads?
A. Check what the campaign is optimising for before switching it off. If it is optimising for page views or form opens rather than real enquiries, the algorithm is finding the wrong people because that is what it was told to do. That is a setup problem, not a verdict on the platform.areness campaigns, and video content that builds trust, but it should not be the primary lead generation channel for most practice areas.