Ask five marketers how much Google Ads cost and you will get five versions of “it depends,” which is the single least helpful sentence in our entire industry. It is also true. So instead of repeating it, we are going to take it apart and show you exactly what it depends on, with real benchmark data, the math that actually matters, and a calculator you can push around yourself.

Here is the short version before we get into it. Across 13,474 US search campaigns analyzed by WordStream’s 2026 Google Ads benchmark study, the average cost per click landed at $5.42 and the average cost per lead came in at $66.69. Your number will be different. This post is about why, and about the number you should be watching instead.

What Google Ads actually cost in 2026

Google Ads cost whatever you tell them to cost, up to a ceiling you set, and then the auction decides how far that money goes. There is no license fee, no minimum contract, and no price list. You set an average daily budget, Google spends against it, and you pay when someone clicks.

The benchmark numbers give you a starting frame. The 2026 WordStream and LocaliQ study, covering April 2025 through March 2026, put the all-industry median at $5.42 per click, a 6.64% click-through rate, an 8.18% conversion rate, and $66.69 per lead.

One genuinely encouraging note from that report: “for the first time in five years, overall average cost per lead in Google and Microsoft Ads has actually gone down.” Travel dropped 39.35% year over year, beauty and personal care dropped 34.95%.

Budgets themselves have a hard edge worth knowing. Google confirms you will never pay more than 30.4 times your average daily budget in a month, though daily spend can run up to twice your daily budget on a busy day. So a $50 daily budget is a $1,520 monthly ceiling, not a $50 guarantee every single day.

How the Google Ads auction decides your price

Every time someone searches, Google runs an auction in the time it takes the page to load, and the winner is not simply whoever bid the most. Google describes it as “a lightning-fast ad auction, that takes place every time someone searches on Google or visits a site that shows ads.”

Picture a bake sale where the judge cares about two things: how much you paid for your entry ticket, and whether your pie is any good. A mediocre pie with a huge ticket can beat a great pie with a tiny one, but a great pie with a decent ticket beats both, and pays less for the privilege.

That is roughly Ad Rank, which Google defines as “a set of values that are used to determine whether your ads are eligible to show and if eligible, where on the page your ads are shown.” It blends your bid, your ad and landing page quality, Ad Rank thresholds, auction competitiveness, and context like device, location, and time of day.

The part people miss is the pricing rule. Google states it plainly: “When you set your bid, you’re telling Google Ads the maximum amount you’re willing to pay for a click on your ad. How much you actually end up paying is often less.”

Translation: your max bid is a ceiling, not a price tag. Better ads and better landing pages let you hold a good position while paying under that ceiling. That is the whole reason two businesses in the same town, bidding on the same keyword, can pay wildly different amounts for the same click.

Quality Score, or why your landing page is a pricing decision

Quality Score is Google’s 1 to 10 report card on how relevant your ads and landing pages are, and improving it is the closest thing to a legal discount on paid search. Google is careful about how it frames this, calling Quality Score “a diagnostic tool meant to give you a sense of how well your ad quality compares to other advertisers,” and noting that “Quality Score is not an input in the ad auction.”

Read that second sentence twice, because it trips people up. The score itself is not plugged into the auction. The three things it measures absolutely are, and Google names them directly in its documentation:

• Expected clickthrough rate: “The likelihood that your ad will be clicked when shown.”
• Ad relevance: “How closely your ad matches the intent behind a user’s search.”
• Landing page experience: “How relevant and useful your landing page is to people who click your ad.”

Those three inputs are the quality half of Ad Rank, so a weak score is a symptom telling you your cost per click is higher than it needs to be.

The practical takeaway is unglamorous but real. If your ad promises emergency drain cleaning and the click drops someone on your homepage, where they have to hunt for the service, you are paying a quality tax on every single click. Building a page that matches the promise is not a nice-to-have. It is a line item that pays for itself.

Average cost per click by industry

Cost per click swings by more than 5x depending on what you sell, and local service businesses sit at the expensive end. Legal, home improvement, and dental all clear $8 a click, because one closed job is worth thousands and everyone bidding knows it.

Here is the 2026 benchmark data for the verticals most of our clients live in, with conversion rate and cost per lead alongside, since a click number alone tells you almost nothing.

Notice the rows that break the pattern. Auto repair has a cheap-ish click and a 15.51% conversion rate, so leads cost $29.96. Apparel has a cheaper click than dental and a cost per lead that is $24 higher, because only 4.50% of those clicks turn into anything.

Year over year, the pressure is mostly on the front end. Legal CPCs rose from $8.58 in the 2025 study to $9.87 in 2026, while conversion rates barely moved. Clicks get more expensive faster than pages get better, which is exactly why the page deserves your attention.

The number that actually matters: cost per customer

Cost per click is a vanity metric with a spreadsheet job. The number that decides whether Google Ads work for you is cost per acquired customer, and it is the only one your bank account recognizes.

Run it out. An HVAC company pays $40 for a click on “emergency AC repair near me.” Feels outrageous. But if 10% of those clicks become leads and half of those leads book a job averaging $1,200, that is 20 clicks, $800 of spend, one job. Fine, not thrilling.

Now change one thing: the job is a $12,000 system replacement. Same $800. Suddenly that $40 click is one of the cheapest things you have ever bought.

Flip it the other way. A $2 click sounds like a steal until 500 of them produce zero leads, at which point you have spent $1,000 on nothing and learned that cheap traffic is not the same as relevant traffic. A click you do not want is expensive at any price.

Three things set your real cost per customer, and only one of them is set by Google:

• Cost per click: the auction’s call, influenced by your quality.
• Landing page conversion rate: entirely yours.
• Lead-to-sale close rate: entirely yours, and usually the most neglected.

Doubling your conversion rate halves your cost per lead without touching your bids, which is why we usually look at the page before we look at the account.

Run your own numbers

Benchmarks describe an average business, and there is no such thing. So rather than asking you to do algebra in your head, we built a calculator: set your monthly budget, your industry’s typical cost per click, and your landing page conversion rate, and it returns estimated clicks, estimated leads, and your estimated cost per lead.

A few things to watch as you move the sliders.

First, drag the conversion rate. Moving from 3% to 6% cuts your cost per lead in half at identical spend. No new budget, no bidding strategy, just a better page. That is almost always the cheapest improvement available.

Second, drag the budget down until the lead count gets small. When your monthly budget only buys eight or ten leads, you are not running a campaign, you are running a coin flip. There is not enough data to tell a good keyword from a lucky one.

Third, take your estimated cost per lead and divide it by the share of leads you actually close. If you close one in four, your true cost per customer is four times that number. Compare that to your average job value or customer lifetime value. If it clears comfortably, spend more. If it does not, the problem is rarely the bids.

The costs nobody puts in the pitch

Ad spend is the cost everyone plans for. The other four are the ones that turn a $1,500 budget into a $2,900 month, so let’s put them on the table honestly.

That last row is the one that surprises people most, so it gets its own section.

The learning period, and the minimum budget that makes sense

New campaigns are not efficient on day one, and that inefficiency is a real cost you should budget for rather than panic about. Google says automated bid strategies need calibration time: “It can take up to around 50 conversion events or 3 conversion cycles for the bid strategy to calibrate to the new objective.”

Think of it like a new hire. Week one, they are slow and they ask a lot of questions. You do not fire them on Friday. You pay for the ramp because you want the version of them that exists in month three.

So the minimum viable budget is not a magic dollar figure, it is a math problem: whatever it costs you to buy roughly 50 conversions in a reasonable window. A dental practice at a $72.97 cost per lead needs something in the neighborhood of $3,600 to get there. An auto repair shop at $29.96 needs closer to $1,500.

Below that line, you are not underfunded so much as flying blind. Every keyword has three clicks and one conversion, nothing is statistically distinguishable from anything else, and the account never improves because there is nothing to learn from. Spending $300 a month for a year usually teaches you less than spending $3,000 for a quarter.

If the real number is out of reach today, narrow the campaign instead of starving it. Fewer keywords, tighter geography, one service. A small budget concentrated on your highest-value job type can absolutely work. A small budget sprayed across forty keywords never does.

Who should not run Google Ads

Plenty of businesses should not be in the auction right now, and we would rather say so than take the retainer. Google Ads amplify what you already have, which means they amplify the gaps too.

Here are the situations where we tell people to fix something else first:

• No clear conversion action. If your site has no form, no trackable phone number, and no booking flow, you are buying traffic you cannot measure. Fix the destination before you buy the road.
• No fast lead response. This one is not a preference, it is measured. Harvard Business Review’s study of 1.25 million sales leads found that firms contacting a lead within an hour were “nearly seven times as likely to qualify the lead” as those who waited just one hour longer, and more than 60 times as likely as those who waited a day. If leads sit in an inbox until Monday, you are paying legal-rate prices for leads a competitor will close.
• Margins that cannot absorb the vertical. At $131.63 per lead in legal, a 25% close rate means roughly $527 per signed client. That is a bargain for a personal injury firm and a catastrophe for a $300 service.
• Nothing to say that is different. If your ad, your offer, and your page are indistinguishable from the four competitors above you, the auction will charge you full price for the privilege of blending in.

None of these are permanent disqualifications. They are just the work that has to happen before paid search is a good use of money instead of a fast way to spend it.

Where this leaves you

How much do Google Ads cost? Somewhere between $2 and $40 a click depending on your industry, plus management, plus a page worth sending people to, plus a learning period. But the only cost that should drive your decision is what you pay for a customer measured against what that customer is worth.

If you want a real answer for your business instead of an industry median, talk to our advertising team. We will look at your numbers, tell you what a realistic budget and cost per customer look like, and tell you honestly if we think your money is better spent somewhere else first.

Frequently Asked Questions