Ask ten people who run Google Ads whether you need a landing page, and most will say yes before you finish the question. Build a separate page for your ads, they say, because a homepage has too many links and too many distractions. Some will add a number to make the case: landing pages convert 20 to 40 percent of visitors, websites convert under 10. We heard that exact range in two widely watched videos on the subject while researching this article. Neither one said where it came from.
So we looked at our own accounts. We pulled every Search click from the last 90 days across the 21 Google Ads accounts we manage that had traffic, about 25,000 clicks in all, and split them by where they landed: the homepage, or a deeper page built for a service, an offer or a location.
The homepage did not lose. It was a near tie.
That does not mean landing pages are a waste of money. It means the question most owners ask, "homepage or landing page," is the wrong one. The page type was never what decided the result. Whether the page answered the search was. Here is what we found, how Google grades the same thing from its side, and how to tell in about 20 minutes whether your business needs a new page or already has the right one.
We Checked 25,000 Google Ads Clicks. The Homepage Didn't Lose.
Between August 13 and November 11, 2026, the Search campaigns in those 21 accounts sent 8,956 clicks to a homepage. Those clicks converted at 4.9 percent. The other 16,538 clicks went to deeper pages, and they converted at 4.7 percent.
Before anyone quotes that as "homepages convert better," here is what it is and is not. It is real account data, pulled from Google Ads' own landing page report. It is not a controlled test. Each account counts its own conversions (a form, a call, a purchase), and different keywords sent people to different pages, so the two groups were not shown the same searches. What it does show is that the gap people describe, a landing page converting four or five times better than a website, did not appear anywhere in this data.
Three accounts let us look closer, because they ran both kinds of page at the same time.
A metal building manufacturer selling in eight states. It has a separate website for each state, and it also has dedicated landing pages we built for its ads. The built landing pages took 3,023 clicks and converted at 6.29 percent. The state websites' homepages took 2,472 clicks and converted at 6.37 percent. That is the same number. Both kinds of page did the same job: someone searched for metal buildings in Texas and landed on a page that was, from the first screen, about metal buildings in Texas.
A chiropractic clinic. Its homepage converted at 6.34 percent on 447 clicks. A dedicated page for one of its locations converted at 6.62 percent on 151 clicks. Close enough that a few more bookings either way would flip it.
A youth debate program. Here the gap was real. Its homepage converted at 3.05 percent on 713 clicks. Its free-class page converted at 6.68 percent on 641 clicks, more than double. The free-class page did not win because it was a landing page. It won because the ads that sent people there offered a free class, and the page was about exactly that, with the sign-up right there. The homepage had to serve parents comparing programs, parents checking schedules and parents looking for tournament results, all at once.
Put the three side by side and the pattern is plain. When the homepage already answered the search, a separate page added nothing. When the ad made a specific offer the homepage did not lead with, the page that matched the offer won by a wide margin.
The Searcher Asked One Question. Your First Screen Has to Answer It.
Every Google Ads click is a short conversation with three lines in it. The person types a search. Your ad answers with a promise. Then they land on your page and check whether the promise holds up. If the first screen they see, before any scrolling, continues that conversation, they keep going. If it changes the subject, many of them leave, and you paid for the click either way.
Google says the same thing in its own advice to advertisers. Its help page on landing pages tells you to choose a page "that closely matches your ad and keywords," and to make the page "mirror the call-to-action in your ad text." Its example: if your ad encourages people to sign up for a free tour, feature a sign-up form prominently on the page. That is the debate program's free-class page, almost word for word.
So the test for any page you send ad traffic to is not "is this a landing page." It is four questions, in order:
- What did they search? Look at the actual searches in your search terms report, not just your keyword list. "Termite inspection near me" and "how to get rid of ants" are two different people.
- What did the ad promise? A service, a price, a speed ("same day"), an offer ("free estimate"), a place ("serving Tampa").
- Does the first screen say that back to them? The headline should name the thing they searched for, in close to their words, and the promise from the ad should be visible without scrolling, on a phone.
- Is the next step obvious? If the ad said "call now," the phone number should be the biggest thing on the screen. If it said "book a free class," the form should be right there.
Picture a pest control company in Tampa, Florida. Its ad for "termite inspection" promises a free inspection within 48 hours. If the click lands on a homepage that opens with a family relaxing in a backyard and the words "Tampa's Pest Control Experts," the person who just found wings on a windowsill has to work out whether this company does termites at all, and whether the inspection is really free. If the click lands on the company's termite page, which opens with "Free Termite Inspection in Tampa. Booked Within 48 Hours." and a button to book it, the conversation continues. Same company, same ad, same budget. One page answers the question and one does not.
The checklist for building the first screen itself, the headline, the one button, the proof next to it, is something we have covered in detail in You Paid for the Click. Your Hero Has 5 Seconds to Keep It. This article is about the decision that comes before that one: which page each ad should send people to, and whether you need to build anything at all.
Your Homepage Answers One Search and Dodges the Other Nine
A homepage has to speak to everyone who might arrive: new customers, existing customers, job applicants, people checking your hours. That is why it is usually a good answer to one kind of search and a weak answer to most of the others. The question is which searches your account is actually paying for.
Your homepage is often enough when:
- You sell one main service. A business that only does window tinting, or only does dog grooming, has a homepage that already is the service page. Building a separate page mostly duplicates it.
- The search is for your business name. Someone who typed your name wants you, not a pitch. Your homepage is the right answer, and a stripped-down landing page with no navigation can actually frustrate them.
- The site is already built around the search. The metal building manufacturer's state websites are the clearest example. Each one is a homepage, and each one is entirely about one state's buildings. That is why they converted as well as the pages we built.
Your homepage usually fails when:
- You sell several services and the ad is about one of them. A plumber's homepage that leads with water heaters will lose the person searching for a clogged drain, even if drain cleaning is listed in the menu. They will not go looking.
- The ad makes a specific offer. A free inspection, a price, a seasonal deal, a free class. If the offer is in the ad and not on the first screen, the person thinks they clicked the wrong thing. This is the debate program's lesson.
- The search is urgent. A burst pipe at 10 p.m. needs a phone number and "available now," not a slider of company photos.
- The phone number or form is buried. Many homepages put the contact button in the top corner and the form at the very bottom. That works for browsers. It loses people who arrived ready to act.
The practical rule that falls out of this is simple: one page per promise, not one page per keyword. If ten keywords in an ad group all mean "I think I have termites," they need one page that answers that. If one ad group is about termites and another is about mosquito treatment, they need two pages, and the second one may well be a page you already have. Most small business websites already have a page for each main service. The fix is often just pointing the ad at it instead of at the homepage.
This connects directly to how you choose keywords in the first place. If your list is full of searches from people ready to hire, which is where a small budget should start (we walk through that in How Do You Choose Keywords for Google Ads?), those people arrive wanting a specific job done. They are exactly the people a generic homepage loses.
Google Grades Your Page Too. The Grade Is Hiding in Your Keyword Report.
You do not have to guess how Google sees your page. It rates it, and the rating sits one click away in your account.
Go to Keywords in the Campaigns menu, open the columns icon, and add the Quality Score columns. One of them is called Landing Page Exp., short for landing page experience. Google defines it as how relevant and useful your landing page is to people who click your ad. Each keyword gets one of three ratings: Above average, Average or Below average. Google compares you with other advertisers whose ads showed for the exact same search over the last 90 days, so the grade is relative. "Below average" means competitors' pages are answering that search better than yours.
What Google says it looks at is worth reading, because it lines up with everything above. Its Ad Rank page lists "the usefulness and relevance of your ad and landing page, the expectations users have about your landing page based on the clicked ad creative," and the ease of navigating the page. In plain English: does the page match what the ad promised, and can a person get what they came for without hunting?
There are two things to keep straight about what this grade does.
First, it affects what you pay, but indirectly. Google is clear that the 1 to 10 Quality Score number itself "is not an input in the ad auction." It is a diagnostic. The page quality underneath it, though, is part of how Google ranks your ad each time it competes, and Google's own wording is that "higher quality ads can often lead to lower CPCs." So a page that matches the search can help you pay less per click, but you will not see that as a neat line from the Quality Score column to your bill.
Second, the grade points at a problem. It does not decide anything. We have seen a keyword with a Quality Score of 3 produce more phone calls than anything else in its account, which is the whole argument of Your Worst Quality Score Might Be Your Best Keyword. A "Below average" landing page rating on a keyword that brings in jobs is a reason to look at the page, not a reason to pause the keyword. A "Below average" rating on a keyword that brings in nothing is a strong hint about why.
There is also a floor below the grade, and it is a hard one. Google's destination requirements policy can stop an ad entirely if the page does not work, if the page sits on a different domain from the one shown in the ad, if Google's crawler cannot read it, or if the page is "unnecessarily difficult or frustrating to navigate." Those are rare on a normal business website. They show up most often after a website rebuild changes page addresses and nobody updates the ads, which is worth checking every time your site changes.
Your Ad Promises One Thing and Your Homepage Says Another. That Gap Is What a Landing Page Fixes.
So when is a dedicated landing page actually worth building? When there is a gap between what the ad promises and what any existing page says, and the gap cannot be closed by pointing the ad somewhere else on your site.
Build a page when:
- The ad carries an offer your site does not feature. A free first class, a free termite inspection, a winter tune-up special. The offer needs a page that opens with it.
- You sell a service your site barely covers. If drain cleaning is a line in a bulleted list on your services page, people searching for drain cleaning need more than a line.
- You serve an area your site does not name. A company based in one city running ads in three neighboring towns will do better when the page says it serves those towns.
- You cannot easily change your website. If every edit means waiting weeks on a web developer, a separate page you control can be the faster fix. This is a common, perfectly good reason.
- You want to test one change at a time. A dedicated page lets you change a headline or an offer without touching the site your existing customers use.
Hold off when:
- Your conversion tracking is not counting real leads yet. If you cannot see which clicks turn into calls and forms, you cannot tell whether the new page did better than the old one. Fix the counting first.
- Your budget cannot tell the pages apart. A $1,500 a month account may buy only a few dozen clicks a month for any one service. Split that across four new pages and none of them gets enough traffic to learn anything for months. Start with one page for your most important service, and send everything else to the best existing page.
- The real problem is the traffic. If the wrong people are clicking, a new page will not convert them. We covered that failure in Your Page Converts Fine. You're Sending It the Wrong People. Check your search terms before you spend on design.
When you do build one, keep it about one thing. The page should open with the searcher's words and the ad's promise, put the phone number or form on the first screen on a phone, and show one piece of real proof near the button, such as your review rating and count. It does not need fancy design. The metal building pages and the state websites converted the same because both were plainly about the thing being searched for, not because either was beautiful.
Your Homepage Already Answers Some Google Searches. The Landing Pages Report Tells You Which.
Before you pay anyone to build a page, spend 20 minutes finding out where your ad money is landing now. Google has a report for exactly this.
In Google Ads, click Campaigns, then Insights & reports, then Landing pages. You will see every page your ads sent people to, with clicks and conversions for each. Set the date range to the last 90 days so small numbers have time to add up.
Then look for three things:
- Pages getting clicks that should not be. Your About page, a blog post, a careers page. If AI Max or Performance Max is turned on, a feature called final URL expansion lets Google send clicks to any page on your site it thinks will convert, and it does pick pages like these. We explain how to check and switch it off in Google Switched Your Search Campaign to AI Max on September 1. You Never Said Yes.
- A homepage carrying most of the clicks in an account that sells several services. That usually means every ad group points at the homepage by default. Check which service each ad group is about, and whether a page for that service already exists.
- Two pages getting traffic for the same service at very different conversion rates. That is your own free test result, the way the debate program's free-class page showed up against its homepage. Send more of the traffic to the page that is winning, and check it again in a month.
Then open your top three pages on your phone, next to the ad that sends people there, and ask the four questions from earlier. What did they search? What did the ad promise? Does the first screen say it back? Is the next step obvious? Most owners find at least one ad pointed at the wrong page, and fixing that costs nothing.
If the answer is that your site genuinely has no page that answers your most important search, that is when a landing page earns its cost. If you would rather have someone run that check and build the page that is missing, it is exactly what we do every day with the Google Ads accounts we manage. And if you would rather do it yourself, you now have the report, the four questions and the numbers to judge the result.
