Landing Page Statistics: Conversion Benchmarks, Form Length Studies and Testing Odds, With Their Fine Print

Fourteen landing page figures and the outfit that measured each

Every figure below was opened on the publisher's own page in October 2026. The last column says who was measured and whether the publisher sells a product the number flatters, which in this field is nearly everyone. Two rows are older than 2024 and flagged; they stay because nothing newer measures the same thing. The rest of the section lives on the marketing statistics hub.

Six point six percent, from pages that converted at least once

Unbounce's 2024 Conversion Benchmark Report is the most quoted landing page figure in circulation, and it is a serious piece of work: more than 41,000 pages, 464 million unique visitors and 57 million conversions between July 23, 2023 and July 23, 2024. The median across all industries was 6.6%, and medians in its nine industries ran from 3.8% to 12.3%.

Read the methodology before you quote it. Pages with fewer than 50 visitors or no conversions at all were excluded, so the median describes landing pages that worked at least a little. A page nobody converted on, which describes plenty of small business pages, is not in the sample. Industries were assigned by machine learning, and groups with fewer than 400 pages were dropped.

The earlier edition shows how much the method matters. Unbounce's 2020 report, from 34,000 pages, put the median at 3.2% and the average at 9.7%. A few very high converters drag an average up, which is why medians are the honest number. The 2020 page does not say whether zero-conversion pages were excluded, so do not read the jump from 3.2% to 6.6% as landing pages doubling in quality. It is two samples, built two ways.

Ad clicks convert on their own scale, and Google keeps the ruler

Paid search conversion rates come from a different counter. LocaliQ, drawing on thousands of its customers' Google Ads and Microsoft Ads campaigns, puts the 2026 average at 8.18%. The spread by industry is the more useful part: Finance and Insurance at 2.64%, Furniture at 2.99%, Automotive Repair, Service and Parts at 15.51% and Animals and Pets at 16.22%. A car repair shop and an insurance agent should not share a target. The cost side of those campaigns is on PPC statistics.

Google's own definition explains why ad platform rates run high. Conversion rate is conversions divided by ad interactions in the same period, and if an account counts every conversion or tracks several actions, the rate can pass 100%, because one click can produce more than one conversion. A page with 50 calls from 1,000 clicks converts at 5%, unless the account also counts a page view as a conversion, in which case it converts at whatever flatters the report.

Unbounce splits its own data by traffic source, as averages rather than medians: email visitors converted at 19.3%, paid social at 12% and paid search at 10.9%. Email audiences already know the sender, which is the argument for building a list rather than renting attention forever. More on email performance is on email marketing statistics.

Form length studies disagree, and each one has a reason

Ask how many form fields a landing page should have and you will hear that every field costs conversions. The research is less tidy than the advice.

• HubSpot analyzed more than 40,000 customer landing pages and reported that conversion fell only slightly as fields were added. Single-line text fields made little difference; multiple text areas and dropdowns had the strongest negative effect. A companion post found lower conversion on pages that asked for age, mentioned phone numbers or asked for location, street address most of all. The posts carry no original date and give no figures in the text, so treat them as direction, not measurement.
• Zuko, which sells form analytics, says in a post last updated December 2025 that 34% of people who start a form do not finish, and that the number of fields has almost no effect on completion. Its benchmark page lists 739 forms, 93 million form views, 60 million starts and 38 million completions, with no date range given.
• Baymard Institute measures checkout forms on large retail sites. Its field counts and its estimate of what fixing them is worth are on our ecommerce website design guide; they describe checkouts, not lead forms.

The disagreement is less mysterious than it looks. Zuko measures people who already started typing, who are committed. HubSpot measured conversion across every visitor to the page, including those who never touched the form. A long form scares people off at the door; once they are inside, length matters less than confusing fields, error messages and requests for information that feels too personal.

Reading habits: the fold, the 20% and the grade-school sentence

Nielsen Norman Group's 2018 eyetracking study, with 120 participants on desktop screens, found people spent about 57% of their viewing time above the fold and 74% in the first two screenfuls. In 2010 the same firm's data had 80% above the fold. People scroll more than they used to, and they still look hardest at the top.

The famous claim that visitors read about 20% of the words on a page comes from the same firm, in May 2008, and rests on 25 users' browsing logged in 2005. It is the oldest number on this page and still the best known. Use it as a reminder that a page is scanned, not as a measurement of your customers.

Unbounce's newer data points the same way. Pages written at a 5th to 7th grade reading level converted at 11.1%, against 5.3% for professional-level writing, and word count, reading time and long words all correlated negatively with conversion. Financial services was the one industry where more complex language helped. 83% of the visits Unbounce analyzed came from mobile devices, so the fold is a phone screen. The layout that follows from this is in the best website layout for lead generation, and the words in copywriting.

Most A/B tests lose, says the company that sells A/B tests

Optimizely analyzed 127,000 experiments run on its platform between 2018 and 2023 and published the results in November 2023. Its summary says only 12% of experiments produce a statistically significant improvement on the primary metric. That is not a scandal. It is the reason testing exists: most ideas that look obvious in a meeting do nothing, and a few do harm.

The background is getting harder, too. Contentsquare's 2026 Digital Experience Benchmark, covering more than 6,500 websites and 99 billion sessions, reports that conversion rates fell 5.1% from Q4 2024 to Q4 2025, while visits fell 3.8% and engagement such as time on site and scroll per visit fell 10%. Fewer visitors, looking for less time, means each page has to do more with what arrives. Speed is part of that, and its numbers are on page speed statistics.

A test that loses is information. A redesign nobody tested is a rumor with a launch date.

How to read these numbers before one lands in a proposal

• Median or average. Unbounce's 2020 report put the median at 3.2% and the average at 9.7%. Ask which one a quoted figure is before you compare yourself with it.
• Who was left out. A benchmark that drops pages with no conversions describes the survivors. Your page, if it is struggling, may not be in the sample at all.
• What counts as a conversion. A click to a second page, a form, a call and a purchase are all conversions somewhere. Google Ads lets one click count several times.
• Who sells the cure. Unbounce sells landing pages, LocaliQ sells ad management, Zuko sells form analytics, Optimizely sells testing. Their data is real; their framing favors the product.
• How old. The reading figure is from 2008 and the fold study from 2018. Habits move, especially on phones.
• What we left out. A widely shared HubSpot figure says personalized calls to action perform 202% better than basic ones. HubSpot's page gives no year for the study, so it is not in our table.

Our rules for every statistic in this section are on the marketing statistics hub, and the way these numbers turn into page design is on the landing page guide.

Your own conversion rate, measured in an afternoon

1. 1

   Decide what a conversion is

   For a service business, it is a call over a minimum length, a form submitted, or a booking. Not a page view, not a scroll.

2. 2

   Count calls as well as forms

   Many local customers phone instead of typing. Without call tracking, a page can look dead while the phone rings. Conversion tracking setup wires both into Google Ads and Analytics.

3. 3

   Divide by the right visits

   Use visits to that page from the campaign or channel that sent them, over at least a full month. Mixing ad traffic with organic traffic blurs both.

4. 4

   Compare with yourself first

   Your page against your page last quarter tells you more than any industry median. Use the benchmarks above only to judge whether a gap is large.

5. 5

   Change one big thing

   Headline, offer, form or proof. Run it for whole weeks, then judge by booked jobs rather than form fills.

If the numbers say the page leaks, our free website audit checks speed, mobile, SEO and content and emails a report with fixes, and the conversion problems guide sorts the usual causes. Ongoing testing is the work of conversion rate optimization. And if the page exists to catch ad clicks, DataCram's PPC management builds the first custom landing page when a program starts, with call and form tracking live before a dollar is spent; the minimum ad budget is $1,000 a month, billed by Google to your own account, and management is month to month after a free audit.