Google Ads vs Facebook Ads: Which One Should You Run First?

“Which is better, Google Ads or Facebook Ads?” is the wrong question, and I get asked it every week. They do different jobs, and one of those jobs is the one your business needs done first.

Google Ads catches people who are already looking for what you sell. Meta ads find people who are not looking yet and give them a reason to. That one difference decides what the ad looks like, what it costs, how fast it works, and what you need ready before you spend a dollar.

The 100 Google Ads accounts I can see spent $12.5M in the last 90 days. Their numbers are in this post, next to the published Meta benchmarks, so you see both bills before you pick.

The difference in one sentence

Google Ads is demand capture. Meta is demand creation.

On Google, the person types “emergency plumber near me” or “crm for small accounting firms”, and your ad shows because of what they typed . You did not have to convince them they have a problem. They told you.

On Meta, nobody typed anything. The ad appears in a feed between a friend’s holiday photos and a video of a dog, and it has to earn attention it was not asked for. The targeting is who the person is and what they have shown interest in , not what they want right now.

That sounds like an argument for Google, and for a lot of businesses it is. But demand capture has a ceiling: you cannot capture demand that does not exist. If nobody is searching for your product, because it is new or because people do not know the category has a name, Google has nothing to catch. Meta can create the search that Google then captures. Plenty of accounts I run have both, in that order.

Side by side

The table is the whole argument. Every line on the left is about a person with a problem in their hand. Every line on the right is about earning a moment from someone who was doing something else.

Where the money goes when both are on the table

Google sells feed ads too. Demand Gen runs in YouTube, Discover and Gmail feeds, and it is the closest thing Google has to a Meta placement. So the 100 accounts I can see are a fair test of what advertisers do when both kinds of ad sit inside one platform.

Across the 75 US accounts, Shopping took 46% of the $12.5M and Search took 34%. Performance Max, which mixes both, took 13%.

The feed products, Demand Gen and YouTube together, took 7%. Display got $15k, which rounds to nothing.

That split is not an accident of my client list. It is what happens when a manager can see the cost of every search term next to the cost of every feed impression. The money goes to the ads that show up when somebody is looking, and the feed gets what is left after those are saturated. That is the order this post recommends, and it is the order the spend already follows.

What a Search click actually costs, from accounts I can see

The most common question after “which platform” is “what will it cost me”. Every benchmark post answers with an average. I would rather show you the spread.

I took the 53 US accounts with at least $1,000 of Search spend in the 90 days to 28 August 2026 and worked out the cost per click, cost per conversion and conversion rate for each one. No client is named. The conversions are whatever each account counts, which is a real caveat and one I come back to below.

The median cost per click is $4.34. The middle half of accounts sit between $2.19 and $5.84. The median cost per conversion is $40, and the middle half run from $22 to $103. The median conversion rate is 7.1%.

What the medians hide is how wide the spread is. Six accounts pay under a dollar a click. Two pay over twelve. The biggest single group, 14 of the 53, sit between $5 and $8, and those are the service businesses with a deadline: the plumbers, the lawyers, the clinics.

My wife’s travel planning business sits at the cheap end. Her qualified leads cost around $7 each on Google, she closes about 1 in 6, and the budget is £1,500 a month. That account is the one I learned Google Ads on , and it is the reason I tell small businesses to check the bid before they assume Google is out of reach.

The published benchmarks, with the warning attached

I said earlier that averages are averages of businesses that are not yours. That is still true. But the published numbers are useful for one thing: they show that the cheap click and the cheap lead are not on the same platform.

LocaliQ’s search benchmarks put the average Google search CPC at $5.42 and the average cost per lead at $66.69. Their Facebook benchmarks put the average Meta CPC for lead campaigns at $1.92 and the cost per lead at $27.66. Both come from the WordStream and LocaliQ customer base, and both are updated through 2025 and 2026.

Read the legal row. A Meta lead at $18 against a Google lead at $131 looks like a seven-to-one win for Meta. It is not, and every lawyer who has run both knows why. The Meta lead is a person who tapped a form in a feed. The Google lead searched “dui lawyer near me” at 2am. One of those becomes a case. The table measures form fills, not clients.

The dentist row is the other lesson. Meta CPC is $9.78 there, higher than Google’s $8.00, because dental practices have piled into Meta and bid the feed up. The cheap platform is only cheap until your competitors find it.

Meta’s own reporting has a bias the benchmarks cannot show you. Haus, which runs incrementality tests for brands, looked at 640 of them and found Meta drove about a 19% lift to the brand’s primary KPI on average , and that for advertisers on click-only attribution Meta under-reports by about 15%. So Meta is often better than its click numbers and worse than its view numbers. The CRM is the number to read.

When Google Ads should come first

Run Google first when the search already exists. That is most local services, most B2B where the buyer knows the problem’s name, and most ecommerce where people compare products before buying.

Services with a deadline. A burst pipe, a lawyer after an arrest, a dentist for a broken tooth. Nobody discovers these needs from a feed. They search, they call the first credible result, and the click is worth a lot because the job is worth a lot. This is why the most expensive keywords on Google are almost all services with a deadline. It is also why I run Google Ads for plumbers, law firms and clinics rather than Meta.

B2B where the problem has a name. “Payroll software for restaurants” is a search. The person typing it has a budget and a timeline. Meta can reach the same restaurant owner, but it reaches them on a Sunday evening when they are not thinking about payroll.

Ecommerce with product search. If people search for the product by name or category, Shopping ads put the product, the price and the image in front of them at the moment they compare. Meta gets the same product in front of people who were not comparing anything. The 46% of spend that went to Shopping in the accounts above is the market voting on this.

The check is simple, and it is the first one below: does Keyword Planner show real monthly searches for what you sell? If it does, Google goes first.

When Meta should come first

Run Meta first when the search does not exist yet, or when the product sells on sight. The Haus line above is the case for it: when Meta works, it works on the whole business, not on a dashboard.

New categories. If you have invented something, nobody is searching for it. There is no demand to capture. Meta can show the product to the people most likely to want it and create the demand. Some of that demand then becomes a Google search, and that is when Google joins.

Products that sell on sight. Clothing, home goods, food, anything where a fifteen-second video does more than a paragraph of text ever could. Search ads cannot show a jacket moving. A feed can.

Low-consideration purchases. If the decision takes one look and costs less than a meal out, the feed is the right place. Nobody researches a phone case.

Retargeting, whichever platform you started on. Once you have traffic, Meta’s remarketing is cheap and it works. Facebook retargeting is a separate post, and it is usually the first Meta campaign I add to a Google account.

Meta first has a cost that Google first does not: creative. A Meta account needs a stream of new images and videos, because a feed audience stops noticing an ad after a few weeks. If you cannot produce creative every month, Meta will get expensive on its own. The Facebook strategies I use cover how to keep that stream going without a studio.

How to decide in an afternoon: three checks

I run these before I recommend a platform to anyone. None of them costs money.

1. Does the search exist?

Open Google Keyword Planner, which is free inside any Google Ads account. Type the ten ways a customer might describe what you sell. Read the monthly searches and the top-of-page bid for each.

If the searches are in the hundreds or thousands a month and the bid is one you can afford against your margin, Google goes first. If the searches are in the tens, or zero, there is nothing to capture. Meta goes first.

This check alone settles it for most businesses. The rest is confirmation.

2. Does the margin cover the cost of a lead?

Take the top-of-page bid from check one. Estimate how many clicks turn into a lead or a sale. The median conversion rate across my 53 Search accounts is 7.1%, so one in fourteen is a fair starting guess for a service site with a decent landing page. For ecommerce, one in fifty. That gives you a cost per lead or per sale on Google before you spend anything.

Now do the same for Meta, using the cost per thousand impressions Ads Manager estimates for your audience and a click rate of around one percent. If the Google number fits inside your margin and the Meta number does not, or the other way round, the platform has chosen itself.

Every number in this check is a guess until you have spent about a thousand dollars. After that, your own account is the only benchmark worth reading. The ROAS calculator and the CPA calculator do the arithmetic if you would rather not.

3. Can you feed the creative?

Google needs a landing page that repeats the search and a few good headlines. It can run for months on the same ad.

Meta needs new images and video every few weeks, or the cost per result climbs as the audience tunes the ad out. If nobody in the business can make or commission creative on a schedule, Meta first is a plan that fails in month two. Start on Google, use the profit to fund the creative, then add Meta.

BUSINESS: ______________   MARGIN PER SALE / VALUE PER LEAD: $______

CHECK 1: DOES THE SEARCH EXIST?  (Keyword Planner, free)
  term                          searches/mo    top-of-page bid
  ____________________________  ___________    $______
  ____________________________  ___________    $______
  ____________________________  ___________    $______
  total searches/mo: ______     (hundreds+ = Google first, tens or zero = Meta first)

CHECK 2: DOES THE MARGIN COVER A LEAD?
  Google: bid $____ x clicks per lead ____ (14 services / 50 ecom) = $____ per lead
  Meta:   CPM $____ / 1000 x 1% click rate = $____ per click x ____ = $____ per lead
  fits inside margin?  Google: Y/N    Meta: Y/N

CHECK 3: CAN YOU FEED THE CREATIVE?
  new images/videos per month you can produce: ____   (Meta needs 4+, Google needs 0)

DECISION: ____________ first.   Add the second when: ____________________________
ONE PRIMARY CONVERSION ACTION, both platforms: ______________________________

Running both without the budget fighting itself

Most businesses I manage end up on both. The order matters, and so does the plumbing between them.

Fix tracking before the second platform. Both platforms will claim the same sale if you let them. Google counts the click, Meta counts the view, and your reports add up to more customers than you have. Before the second platform launches, pick one primary conversion action that means money, set it up the same way on both, and read the CRM rather than either dashboard.

Split the budget by what is proven. When the second platform starts, it gets about a fifth of the budget. It keeps that until its cost per result on the one conversion that matters is within reach of the first platform’s. Then it earns more. A fifty-fifty split on day one is how a profitable Google account becomes an unprofitable two-platform account.

Give each platform its job. Google captures the searches. Meta retargets the visitors Google sent, then prospects for people who look like the customers Google found. That sequence is why the order matters: Meta’s prospecting gets better when it has real customers to model on.

Read them together, not separately. A Meta campaign that shows no direct sales and a brand search campaign whose volume doubled are the same story. How PPC budgets get set and adjusted covers the monthly rhythm. The channel expansion planner is the tool I use to decide whether an account is ready for its second platform, and what has to be true first.

Four mistakes I see every month

Choosing by preference rather than by check one. The founder likes Instagram, so the plumbing company runs Instagram. Nobody scrolling Instagram has a burst pipe.

Launching both at once with no tracking. Two dashboards, both green, one bank account that did not move.

Starting Meta with one image. By week three the audience has seen it forty times. The cost per result climbs and the platform gets blamed for a creative problem.

Treating Google’s broad match like Meta’s audiences. Broad match with no negative list spends like a prospecting campaign with no ceiling. The negative keyword lists are where a Google account’s waste gets stopped.

If you would rather have both run for you

Is Google Ads or Facebook Ads better?

Neither. Google Ads catches people who are already searching for what you sell. Meta ads find people who are not searching yet. If the search exists, Google Ads goes first. If it does not, or the product sells on sight, Meta goes first. Most businesses end up on both, in that order.

Which is cheaper, Google Ads or Facebook Ads?

Per click, Meta is usually cheaper: LocaliQ's benchmarks put the average Meta lead-campaign click at $1.92 against $5.42 on Google search. Per customer, it depends on whether the person was looking. Across 53 US Search accounts I can see, the median cost per click is $4.34 and the median cost per conversion is $40. Run the cost-per-lead check in this post with your own margin.

Should a small business start with Google Ads or Facebook Ads?

Check Keyword Planner first. If people search for what you sell in the hundreds a month, start with Google Ads, because the person is ready and you need almost no creative. If nobody searches for it, or you cannot compete on the bid, start with Meta and plan for a stream of new images and videos.

Can you run Google Ads and Facebook Ads at the same time?

Yes, and most accounts should eventually. Fix tracking first with one primary conversion action on both platforms, give the second platform about a fifth of the budget until it proves its cost per result, and give each platform its job: Google captures searches, Meta retargets the visitors and prospects for lookalikes.

Why do Google Ads and Facebook Ads both claim the same sale?

Google counts the click, Meta counts the click or the view, and neither knows about the other. Read the CRM or the order system, not the dashboards. The total there is the only number that matters, and the split between platforms is decided by what happens when you turn one down.