Noémi arranges wedding flowers in Szeged. She also does funeral work, a weekly subscription bouquet and the occasional workshop. A friend who "does websites" has asked her one question before starting: do you want a landing page or a full website?
She has no idea. Her cousin says a landing page converts better. A forum says one-page sites are bad for Google. A web builder's ad says a full site "builds authority". All three sound confident. None of them mentions what Noémi sells, who finds her, or where they come from.
Noémi and the other people in this chapter are imagined. The studies, policies and statements quoted are real, and every one is linked in the sources.
The trouble is the question itself. "Landing page or website?" sounds like a choice between two products. It is really three smaller questions wearing one coat, and once you pull them apart, the answer is usually obvious. This chapter shows you how, what the evidence does and does not say, and where the popular advice is built on thin air.
Three questions in a trench coat
Before comparing anything, separate the three things people mean when they say "landing page" or "website".
1. How many pages? One, or several. This is the only question the words themselves answer.
2. How many goals? A page can try to do one thing (book a call) or many things for many people (introduce the company, list prices, answer questions, sell a product, recruit staff). The first kind is what marketers call a landing page. The second is what the Nielsen Norman Group calls a homepage: in Huei-Hsin Wang's 2024 principles, the front door of a website and its "most important routing page", the place people return to when they get lost.
3. Where do visitors come from, and what are they thinking when they arrive? Someone who clicked your ad for "emergency plumber" is in a different state of mind from someone who typed your name after a friend's recommendation.
These are independent. A website can contain landing pages. A landing page can be one page of a larger site. A one-page site can have one goal or six. Mixing them up is where most bad advice starts.
What to do: before choosing anything, write three lines. How many different kinds of visitor do I have? What is the one thing I want each to do? Where will they come from? Keep those lines. The rest of the chapter uses them.
Start with where they come from
The single most useful fact for the decision is not how big your business is. It is how people arrive.
Paid traffic is the clearest case. Google Ads rates every ad's landing page experience as one of the three components of Quality Score, alongside expected click-through rate and ad relevance. Google defines it as how relevant and useful your page is to people who click your ad. The ad promised something specific; the page has to deliver it, right there on arrival.
That is what practitioners call message match: the headline of the page repeats the promise of the ad that sent people to it. Unbounce's 2024 Conversion Benchmark Report recommends it, especially for paid search. Think of someone who searched "emergency boiler repair Szeged", clicked an ad that said "Boiler fixed today", and landed on a homepage with a photo of a smiling team and the headline "Quality since 2009". They did not get what was promised. They go back and click the next ad.
Other sources of visitors want different things:
| They arrive from | What is in their head | Send them to |
|---|---|---|
| An ad for one specific thing | "Is this the thing I just clicked for?" | A dedicated landing page that repeats the ad's promise |
| A search for your name or a recommendation | "Are they real? What do they do? Can I trust them?" | A homepage or one-page site that says who you are first |
| A local search such as "florist near me" | "Who is close, open and good?" | Your Google Business Profile first, then your page |
| A search for one of several services | "Do they do this exact thing?" | The page about that service |
| A post, email or QR code you control | Whatever you promised in the post | A landing page matching the post |
Notice that only the first and last rows clearly want a landing page. The middle rows want a site that answers a broader question. That is the first reason the popular claim "landing pages always convert better" falls apart: the pages are doing different jobs, for different people.
On Google Ads specifically: the official destination requirements say your landing page must work properly across common browsers and devices, must lead to the same domain shown in your ad, and must offer something of its own. The insufficient original content policy lists pages that mainly show ads, copy other sources without adding value, or only send users somewhere else. A landing page built from a thin template can get an ad disapproved.
What to do: list your top three sources of visitors, including the ones you plan to pay for. Next to each, write the exact words you expect the visitor to have in mind. Your page's headline should repeat those words.
What the numbers say, and what they do not
You will see conversion statistics quoted at you. Here are the best public ones, with their limits.
These come from Unbounce's 2024 report, covering roughly 41,000 landing pages, 464 million visitors and 57 million conversions over a year. It is a large dataset, with three caveats worth knowing.
First, Unbounce sells landing page software, so the pages in the data were built by people who care about landing pages. Second, a "conversion" means each page owner's own goal: a purchase, a sign-up, a click. Third, the report measures landing pages against each other. It does not compare landing pages with full websites. We found no public study that does so fairly, and you should be suspicious of anyone who quotes one.
What the report does offer are patterns inside landing pages. Pages with more words tended to convert slightly worse (a correlation of −18.6%). Pages written at a 5th to 7th grade reading level converted at 11.1%, about 56% higher than pages written at an 8th or 9th grade level, and more than twice the 5.3% of professional-level writing. Financial services was the one industry where complex language did not hurt. Correlation is not cause (short, plain pages may simply belong to simpler offers), but plain words are a free thing to try. The copy chapter shows how.
A statistic we chose not to repeat
You may have read that companies with 10 to 15 landing pages generate 55% more leads than those with fewer. It is quoted everywhere and attributed to HubSpot. We could not find the original report behind that exact figure; the versions we found disagree on the numbers (some say 31 to 40 pages bring seven times more leads), and all of it is correlation. Companies with many landing pages are usually companies running many campaigns, which is a different thing from landing pages causing leads. We leave it out.
When one page is plenty
The fear Noémi's forum raised, that one-page sites are bad for Google, is old and mostly unfounded. In a December 2021 Search Central office-hours session, Google's John Mueller was asked about it. As Search Engine Journal reported, he said, "I think you can make good one-page sites", and that a small site such as a business with one product or its opening hours does not have to be an authority. He added a sensible warning: plan for growth, so you do not end up cramming everything onto one page.
Google's guidance on helpful content agrees in spirit. It talks about content made for people and the quality of the main content, and it does not make a page count a ranking requirement. What it warns against is producing lots of content on many topics hoping some of it performs.
A single page also fits how people read. NN/g's 2018 eye-tracking study (more than 130,000 fixations from 120 participants) found that visitors spent about 57% of their viewing time above the fold, 74% in the first two screenfuls and 81% in the first three. People do scroll, but the top of the page does most of the work.
Two things follow. A long single page is fine if it is ordered by importance (the section order chapter). And the most important thing, your offer and your action, belongs near the top rather than being saved for a big finish.
Also, a one-page site cuts down on a lot of work that a site with twelve pages creates: twelve pages to keep accurate, twelve places for a price to be out of date. For a business with one clear offer, fewer pages is not laziness. It is quality control.
What to do: if a visitor could understand and act on your offer with a few minutes of scrolling, start with one page. Put the offer and the action in the first screen.
Where one page runs out of road
One page does have a ceiling, and it is not about how Google feels about one-pagers. It is about how people search.
A page can only be one answer. Google's SEO starter guide asks for a title that is unique to the page and accurately describes its contents. A page has one title and one address. That is a natural fit when everyone is looking for the same thing. It is a poor fit when your business has several services that people search for separately.
That last sentence is our reading of how search works, not a Google rule, so treat it as a reasoned judgment. Consider a bookkeeper who also does payroll and VAT filing. People who need payroll help search for payroll help. A single "Accounting services" page can mention payroll in one paragraph, but a page that is about payroll, with its own title and its own examples, is a better answer to that search.
A trap: pages made for search engines
Once people learn that more pages can mean more searches, a bad idea follows: copy the same page fifty times with a different town name. Google's spam policies have a name for it, doorway abuse: pages created to rank for specific, similar searches, including multiple pages for particular regions or cities that all funnel people to one place. A new page is justified by a real, different answer for a real, different visitor, not by a different place name in the headline.
Local businesses: your page is not the only door
If most of your customers find you through "near me" searches, remember that Google's local results work on different grounds. According to the Business Profile help page, local results rely mainly on three factors: relevance, distance and prominence, the last drawing on reviews and information from across the web. Your website is not one of the three named factors. A small local business can win people over with a complete Business Profile and good reviews, and a clear single page for those who tap through.
What to do: list each service you want to be found for. If two or more are things people search for separately, they deserve their own pages sooner or later. If they are all variations of one thing, keep one page.
The quiet advantages: speed and upkeep
Fewer pages with fewer moving parts are easier to make fast, and speed has been measured. In a study commissioned by Google and carried out by fifty-five and Deloitte Digital, "Milliseconds make millions", researchers analysed 30 million+ sessions across 37 brand sites. A 0.1-second faster mobile site was associated with 8.4% higher conversions in retail and 10.1% in travel.
Read that with care. It involved large European and American brands, not your florist's site; Google funded it; and the effect appeared only when every page in the visitor's journey got faster by 0.1 seconds. That last condition is the useful part: a one-page site has one page to speed up. For individual companies, web.dev's collection of case studies reports gains such as Vodafone Italy's 8% increase in sales after improving Largest Contentful Paint by 31%, with the usual note that these are the companies' own figures. The Core Web Vitals chapter shows what to measure.
The same logic applies to upkeep. Every page is a promise that its information is still true. Hours, prices, a retired service, a phone number: the more pages, the more places a wrong fact can hide. Small teams rarely keep twenty pages current.
Short page or long page?
This is a separate question from "one page or many", and people confuse them. A landing page can be a screen and a half or ten screens. What decides?
The honest answer from Unbounce's own guidance is: the size of the commitment. Its template guidance says long pages suit high-cost or high-stakes offers, where the visitor needs more reassurance before deciding, and that simple offers do well with brief pages and a short form. Nobody should write 2,000 words to sell a 9-euro item, and nobody should say "Contact us!" and a stock photo to sell a 9,000-euro wedding package.
The report's finding that fewer words correlate with higher conversion is real, but it is an average across offers. A rule that says "shorter is always better" is as wrong as one that says "longer is always better". The right length is the length at which your visitor's questions have been answered, and no longer. The lead forms chapter covers the other half of the commitment: how many fields to ask for.
Six quick questions. The result is a rule of thumb built from this chapter, not a scientific instrument. Use it as a starting point for a conversation, not a verdict.
Answer all six to see a suggestion.
You can start small and grow
The choice is not permanent. The best path for most small businesses is a staircase: begin at the step that fits, and take the next step when a clear signal says so, not before.
Mueller's "plan for growth" advice from earlier is the thing to keep in mind. Build a one-page site so each section can later become its own page: give each section a clear heading and an anchor, keep section content self-contained, and use plain, descriptive URLs (the SEO starter guide recommends this). Then promoting a section into a page is an afternoon of work instead of a rebuild.
Note the last step. Even a large website keeps landing pages. Whenever you run a campaign, an email or an ad, a page built for that one promise will outperform your homepage for that visitor. So the real picture is rarely "landing page or website". It is "a website, plus landing pages whenever you send traffic with a specific promise".
Four people, four answers
Here is the reasoning applied to four imagined businesses.
Landing page
Gergő buys ads for "emergency plumber" and gets almost all calls from them. One service, one visitor type, one action: call. A single landing page that repeats the ad's promise, shows a phone button, prices and reviews, plus a legal footer. A full site would add pages nobody visits.
One-page site, with a plan
Weddings, funerals, subscriptions and workshops are four different people with four different moods. Start with a one-page site with a short anchor menu and a clear first section for weddings (the biggest earner). When the subscription grows, it gets its own page, and an ad for it gets its own landing page.
Small site, 5 pages
Six treatments, each searched for separately ("sciatica physio", "sports injury physio"), plus a team and booking. One page cannot be the specific answer to each. A small site: home, treatments (with a page each for the two biggest), team, prices, contact. Booking-focused landing pages for any paid campaign.
Full website
Two hundred products, a blog and weekly stock changes. This is shop territory: categories, product pages, search. It also needs landing pages for each seasonal campaign, so the ads do not land on a shelf of 200 products.
Five ways this goes wrong
Sending ads to the homepage
The most common waste of ad money. The homepage tries to serve everyone, so it serves the clicker poorly. Send paid traffic to a page matching the ad.
A landing page with no trust
Stripping a page down is not the same as stripping out proof. Keep reviews, a real name and a phone number. The trust signals chapter lists what people look for, and the legal footer is not optional (see navigation).
A one-page site with no map
Fourteen sections and no anchors, so people cannot jump. Use a short menu or a table of contents, and make anchors land in view.
Many near-identical pages
Fifty "plumber in [town]" pages with the same text. That is the doorway pattern described above, and it makes the whole site look thin.
Choosing by fashion
"Everyone has a landing page now" or "a real business needs a real website" are both fashion. Choose by your sources of visitors, your goals and your content.
What to change tomorrow
Write your three lines
Kinds of visitor, the one action for each, and where each comes from. Everything else follows.
Match the page to the source
One page per promise. If you pay for a click, send it to a page whose headline repeats the ad.
Start with the smallest shape that works
One goal and one audience: a landing page. A few related offers: a one-page site. Separately searched services: a small site.
Make the top of the page do the work
Offer and action in the first screen. About three-quarters of viewing time falls in the first two screens.
Write plainly
Short words, short sentences. Test the page at a lower reading level than you think you need.
Plan the staircase
Headings, anchors and clean URLs now, so a section can become a page later without a rebuild.
Keep landing pages for campaigns, forever
Whatever your site becomes, each paid promise deserves its own page. Then run the ten-minute audit on it before spending a euro.
Noémi did not need a verdict between two products. She needed three lines on a piece of paper, and the paper said: one-page site now, with the wedding section first, a subscription page later, and a landing page the day she buys her first ad.
What the research supports
Strong: ad landing pages are rated by Google on relevance and usefulness; message match is widely recommended by practitioners, and it fits how Google defines relevance. Solid but limited: Unbounce's 2024 benchmark is large and recent but comes from a landing page vendor, and it compares landing pages with one another, not with websites. Google has said one-page sites can be good. Attention concentrates at the top of a page (NN/g, 120 participants). Reasoned, not proven: that separately searched services eventually need separate pages; that thresholds like "weekly content" signal a bigger site. Weak: the often-quoted "10 to 15 landing pages gives 55% more leads", which we could not trace to a primary source.
Sources
- Unbounce (2024). Conversion Benchmark Report. Consulted for the sample (41K+ pages, 464M+ visitors, 57M+ conversions), the 6.6% overall median, the 3.8–12.3% industry range, the word-count, reading-level and message-match findings.
- Unbounce. Long-form sales page template. Consulted for guidance on when long and short pages suit an offer.
- Google Ads Help. About Quality Score. Consulted for the three components and the definition of landing page experience.
- Google Ads Policy Help. Destination requirements and Insufficient original content. Consulted for what ad destinations must do and what gets them disapproved.
- Search Engine Journal (13 December 2021). Google: You don't need to be an authority to rank. Report of John Mueller's 10 December 2021 Search Central office hours. Secondary source for the one-page quote.
- Google Search Central. Creating helpful, reliable, people-first content. Consulted for the content-quality principles.
- Google Search Central. SEO starter guide. Consulted for page titles, URLs and site organization.
- Google Search Central. Spam policies for Google web search. Consulted for the doorway abuse definition and examples.
- Google Business Profile Help. Tips to improve your local ranking on Google. Consulted for relevance, distance and prominence.
- Wang, H.-H. (2024). Homepage design: 5 fundamental principles. NN/g. Consulted for what a homepage is for.
- Fessenden, T. (2018). Scrolling and attention. NN/g. Consulted for the 57%, 74% and 81% viewing-time figures and the study design.
- Google / Deloitte Digital / fifty-five (2020). Milliseconds make millions. web.dev. Consulted for the 0.1-second findings, sample and conditions.
- web.dev. The business impact of Core Web Vitals. Consulted for the Vodafone Italy example. Figures are the companies' own.
Sources checked 19 October 2026. We found no public study that compares landing pages with full websites on equal terms, so the chapter does not claim one converts better than the other. Noémi, Gergő, Dr Aranka, Studio Kovács and the businesses in the examples are fictional; all figures are original illustrations except Figure 2, which is redrawn from NN/g's published numbers.