Landing Page Statistics 2026: Conversion Rate Benchmarks
Sourced landing page statistics: Unbounce’s 6.6% median, conversion rates by industry, traffic source and device, Core Web Vitals and A/B test win rates.
Median landing page conversion rate, 41,000+ pages, Jul 2023–Jul 2024 (Unbounce)
Median conversion rate, US search ad campaigns, Apr 2025–Mar 2026 (WordStream)
Mobile websites with good Core Web Vitals, July 2025 (HTTP Archive)
Experiments with a significant win on the primary metric (Optimizely)
Key Takeaways
What is a good landing page conversion rate in 2026, and what actually moves it? Many of the statistics in circulation are paid-search account figures from 2014, single vendor case studies or retail speed studies from 2017, repeated without their dates or credited to reports that do not exist.
This page collects landing page and conversion figures with their source, sample and period, and labels the few vendor claims that state no method: Unbounce’s 2024 Conversion Benchmark Report, WordStream and LocaliQ’s 2026 ad benchmarks, Contentsquare’s 2026 Digital Experience Benchmark, HTTP Archive’s Web Almanac 2025, StatCounter, WebAIM, Google’s web.dev guidance and case studies, Nielsen Norman Group, Baymard Institute and peer-reviewed experimentation research. Older studies appear only where nothing newer exists, with their year. Figures were checked on October 21, 2026. For help applying them to your own pages, see our web development work.
How to read this page: “Benchmark (platform data)” comes from one company’s own customer or platform data: Unbounce’s landing page figures are medians unless a row says average, and its device and traffic-source figures are averages unless a row says median; WordStream and LocaliQ’s are medians across US ad campaigns, which WordStream calls averages; Contentsquare’s cover whole websites, not landing pages only; Optimizely’s cover experiments run on its platform; Akamai’s 2017 figures come from SOASTA’s retail monitoring data. “Benchmark (agency research)” is Portent’s 2022 study of 20 websites. “Measured (independent dataset)” is HTTP Archive, StatCounter and WebAIM. “Guideline (Google)” is Google’s published Core Web Vitals thresholds. “Case study (company-reported)” is one company’s result reported through Google or Harvard Business Review. “Study (commissioned research)” is Deloitte’s study for Google. “Model estimate” is a figure from a statistical model, not a measured outcome. “Measured (Google research)” is Google’s own test of mobile ad landing pages. “Estimate (average of published studies)” is Baymard Institute’s average of 50 cart abandonment studies. “Research (usability study)” and “Academic research” are Nielsen Norman Group studies and peer-reviewed or university research. “Survey” and “Benchmark (UX audit)” are Baymard Institute’s shopper survey and site reviews. “Vendor claim” is a figure a vendor states about its own product without a published method. Every source is listed once in the Sources section at the end.
Landing Page Conversion Benchmarks by Industry
The most recent large landing page benchmark is Unbounce’s 2024 Conversion Benchmark Report. It covers more than 41,000 landing pages built on Unbounce, 464 million visitors and 57 million conversions between July 2023 and July 2024, and puts the median conversion rate across all industries at 6.6%. Unbounce has not published a 2025 or 2026 edition, so any “2026 Unbounce benchmark” in circulation is not one of its reports.
Industry medians run from 3.8% for SaaS to 12.3% for entertainment, where sweepstakes pages pull the figure up. Subcategories spread even wider: insurance pages convert at 18.2% and investing pages at 3.9% within the same finance industry. Pages with fewer than 50 visitors or no conversions were left out of the data, so these medians describe pages that convert at least some of the time.
Unbounce’s 2020 edition gave a 3.2% median and a 9.7% average for the same kind of pages. The gap between the two is one reason published “average conversion rates” disagree: a small number of very high-converting pages lifts the average well above the typical page. For benchmarking work on your own pages, see ouranalytics and insights services.
| Figure | What it measures | Source |
|---|---|---|
| 6.6% | Median conversion rate across all landing pages built on Unbounce, July 2023 to July 2024 (the latest edition; Unbounce has published no 2025 or 2026 report) | Unbounce, Aug 29, 2024 · data Jul 2023–Jul 2024Benchmark (platform data) |
| 41,000+ | Landing pages in Unbounce’s 2024 benchmark, with 464 million unique visitors and 57 million conversions between July 23, 2023 and July 23, 2024; pages with fewer than 50 visitors or no conversions were excluded | Unbounce, Aug 29, 2024 · data Jul 2023–Jul 2024Benchmark (platform data) |
| 12.3% | Entertainment (sweepstakes, publishing, games and streaming): median landing page conversion rate, the highest of nine industries, pulled up by sweepstakes pages (median 47.5%); landing pages built on Unbounce, July 2023 to July 2024 | Unbounce, Aug 29, 2024 · data Jul 2023–Jul 2024Benchmark (platform data) |
| 8.4% | Education: median landing page conversion rate (online course pages 18.3%, higher education 6.3%); landing pages built on Unbounce, July 2023 to July 2024 | Unbounce, Aug 29, 2024 · data Jul 2023–Jul 2024Benchmark (platform data) |
| 8.3% | Finance and insurance: median landing page conversion rate (insurance 18.2%, investing 3.9%); landing pages built on Unbounce, July 2023 to July 2024 | Unbounce, Aug 29, 2024 · data Jul 2023–Jul 2024Benchmark (platform data) |
| 6.3% | Legal: median landing page conversion rate; landing pages built on Unbounce, July 2023 to July 2024 | Unbounce, Aug 29, 2024 · data Jul 2023–Jul 2024Benchmark (platform data) |
| 6.1% | Commercial and professional services: median landing page conversion rate; landing pages built on Unbounce, July 2023 to July 2024 | Unbounce, Aug 29, 2024 · data Jul 2023–Jul 2024Benchmark (platform data) |
| 5.1% | Health and wellness: median landing page conversion rate (wellness pages 8.2%, medical treatment 5.3%); landing pages built on Unbounce, July 2023 to July 2024 | Unbounce, Aug 29, 2024 · data Jul 2023–Jul 2024Benchmark (platform data) |
| 4.8% | Travel and hospitality: median landing page conversion rate (accommodation pages 3.7%); landing pages built on Unbounce, July 2023 to July 2024 | Unbounce, Aug 29, 2024 · data Jul 2023–Jul 2024Benchmark (platform data) |
| 4.2% | Ecommerce: median landing page conversion rate (food and beverage 7.1%, fashion and beauty 1.3%; one passage of the same page gives 5.2%); landing pages built on Unbounce, July 2023 to July 2024 | Unbounce, Aug 29, 2024 · data Jul 2023–Jul 2024Benchmark (platform data) |
| 3.8% | SaaS: median landing page conversion rate, the lowest of nine industries; landing pages built on Unbounce, July 2023 to July 2024 | Unbounce, Aug 29, 2024 · data Jul 2023–Jul 2024Benchmark (platform data) |
| 3.2% | Median conversion rate across 34,000 landing pages built on Unbounce, October 2018 to October 2019 (186 million visits, 19 million conversions); the same pages averaged 9.7% | Unbounce, Apr 17, 2020 · data Oct 2018–Oct 2019Benchmark (platform data) |
| 25% | Median conversion rate of the top quarter of landing pages in Unbounce’s 2020 benchmark (October 2018 to October 2019) | Unbounce, Apr 17, 2020 · data Oct 2018–Oct 2019Benchmark (platform data) |
Conversion Rates by Traffic Source
Where a visitor comes from changes the conversion rate a great deal. In Unbounce’s 2024 data, visitors arriving from email converted at 19.3% on average, against 12% for paid social and 10.9% for paid search. Among paid social sources, Instagram (17.9%) and Facebook (13%) led, and LinkedIn converted four to five times less often. Unbounce counts conversions, not their value, so a lower rate from a business audience can still be the better result.
Whole-site data, which counts every session rather than visits to a dedicated page, gives lower numbers. Across Contentsquare’s 6,500+ websites in the fourth quarter of 2025, paid search converted at 2.8%, the highest of the paid channels, and returning visitors converted at 2.9% against 1.7% for new ones. For channel-level figures across the whole funnel, see ourconversion rate benchmarks by industry and channel.
| Figure | What it measures | Source |
|---|---|---|
| 19.3% | Average conversion rate of visitors arriving from email, all industries; landing pages built on Unbounce, July 2023 to July 2024 | Unbounce, Aug 29, 2024 · data Jul 2023–Jul 2024Benchmark (platform data) |
| 12% | Average conversion rate of visitors arriving from paid social ads, all industries; landing pages built on Unbounce, July 2023 to July 2024 | Unbounce, Aug 29, 2024 · data Jul 2023–Jul 2024Benchmark (platform data) |
| 10.9% | Average conversion rate of visitors arriving from paid search ads, all industries; landing pages built on Unbounce, July 2023 to July 2024 | Unbounce, Aug 29, 2024 · data Jul 2023–Jul 2024Benchmark (platform data) |
| 11.3% | Average conversion rate of visitors from Google search ads, 41% above Bing and 95% above Yahoo search ads; landing pages built on Unbounce, July 2023 to July 2024 | Unbounce, Aug 29, 2024 · data Jul 2023–Jul 2024Benchmark (platform data) |
| 17.9% | Average conversion rate of visitors from Instagram ads (Facebook ads 13%), the two highest paid social sources; landing pages built on Unbounce, July 2023 to July 2024 | Unbounce, Aug 29, 2024 · data Jul 2023–Jul 2024Benchmark (platform data) |
| 6–9% | Average conversion rate range for visitors from YouTube, TikTok and X ads; landing pages built on Unbounce, July 2023 to July 2024 | Unbounce, Aug 29, 2024 · data Jul 2023–Jul 2024Benchmark (platform data) |
| 4–5x lower | Conversion rate of visitors from LinkedIn ads compared with Facebook and Instagram ads (Unbounce counts conversions, not their value); landing pages built on Unbounce, July 2023 to July 2024 | Unbounce, Aug 29, 2024 · data Jul 2023–Jul 2024Benchmark (platform data) |
| 0.3% | SaaS: median conversion rate of visitors arriving from display ads; landing pages built on Unbounce, July 2023 to July 2024 | Unbounce, Aug 29, 2024 · data Jul 2023–Jul 2024Benchmark (platform data) |
| 2.8% | Conversion rate of paid search visits, the highest of the paid channels, across 6,500+ websites in nine industries (whole-site sessions, not landing pages only), Q4 2025 | Contentsquare, read Oct 21, 2026 · Q4 2025 dataBenchmark (platform data) |
| 2.9% | Conversion rate of returning visitors, against 1.7% for new visitors, across 6,500+ websites (99 billion sessions), Q4 2025 | Contentsquare, read Oct 21, 2026 · Q4 2025 dataBenchmark (platform data) |
Paid Search and Paid Social Conversion Rates
Ad benchmarks measure conversions per ad click, which makes them the nearest public yardstick for paid landing pages. WordStream and LocaliQ’s 2026 benchmarks put the median for US search campaigns at 8.18%, from 13,474 campaigns on Google Ads and Microsoft Ads between April 2025 and March 2026. By industry it runs from 2.64% in finance and insurance to 16.22% in animals and pets. WordStream’s first report, in 2016, gave 2.70%.
For Facebook leads campaigns, the 2026 median was 8.54% across 452 US campaigns. The often-quoted 2.35% is a different measure from a different decade: the median conversion rate of Google Ads accounts in a 2014 WordStream analysis. The post’s introduction calls it a landing page rate, but the analysis measured whole accounts, and WordStream notes that its 11.45% top-10% figure is not for individual landing pages. For cost and click-through data on the same campaigns, see ourGoogle Ads benchmarksandFacebook ads benchmarks.
| Figure | What it measures | Source |
|---|---|---|
| 8.18% | Conversion rate of search ad clicks, all industries, median across 13,474 US search campaigns on Google Ads and Microsoft Ads, April 2025 to March 2026 (WordStream calls its medians averages) | WordStream, May 18, 2026 · Apr 2025–Mar 2026 dataBenchmark (platform data) |
| 16.22% | Animals and pets: conversion rate of search ad clicks, the highest of 23 industries; median across 13,474 US search campaigns on Google Ads and Microsoft Ads, April 2025 to March 2026 | WordStream, May 18, 2026 · Apr 2025–Mar 2026 dataBenchmark (platform data) |
| 13.14% | Education and instruction: conversion rate of search ad clicks; median across 13,474 US search campaigns on Google Ads and Microsoft Ads, April 2025 to March 2026 | WordStream, May 18, 2026 · Apr 2025–Mar 2026 dataBenchmark (platform data) |
| 12.43% | Physicians and surgeons: conversion rate of search ad clicks (dentists 10.67%); median across 13,474 US search campaigns on Google Ads and Microsoft Ads, April 2025 to March 2026 | WordStream, May 18, 2026 · Apr 2025–Mar 2026 dataBenchmark (platform data) |
| 5.55% | Attorneys and legal services: conversion rate of search ad clicks; median across 13,474 US search campaigns on Google Ads and Microsoft Ads, April 2025 to March 2026 | WordStream, May 18, 2026 · Apr 2025–Mar 2026 dataBenchmark (platform data) |
| 3.70% | Real estate: conversion rate of search ad clicks; median across 13,474 US search campaigns on Google Ads and Microsoft Ads, April 2025 to March 2026 | WordStream, May 18, 2026 · Apr 2025–Mar 2026 dataBenchmark (platform data) |
| 2.64% | Finance and insurance: conversion rate of search ad clicks, the lowest of 23 industries (furniture 2.99%, career and employment 3.05%); median across 13,474 US search campaigns on Google Ads and Microsoft Ads, April 2025 to March 2026 | WordStream, May 18, 2026 · Apr 2025–Mar 2026 dataBenchmark (platform data) |
| 8.54% | Conversion rate of Facebook leads-objective campaigns, all industries, median across 452 US campaigns, April 2025 to June 2026 (WordStream does not say how many used a landing page rather than Meta’s lead forms) | WordStream, Sep 14, 2026 · Apr 2025–Jun 2026 dataBenchmark (platform data) |
| 15.87% | Education and instruction: conversion rate of Facebook leads campaigns, the highest industry (automotive for sale lowest at 4.15%), median across 452 US campaigns, April 2025 to June 2026 | WordStream, Sep 14, 2026 · Apr 2025–Jun 2026 dataBenchmark (platform data) |
| 2.70% | Average conversion rate of search ad clicks in WordStream’s first industry benchmark report, 2016, for comparison with 8.18% in 2026 | WordStream, May 18, 2026 · Apr 2025–Mar 2026 dataBenchmark (platform data) |
| 2.35% | Median conversion rate of Google Ads accounts in a 2014 WordStream analysis of “hundreds of accounts” over three months (top 25% at 5.31% or more, top 10% at 11.45%); account-level, not individual landing pages | WordStream, Mar 17, 2014Benchmark (platform data) |
Mobile vs Desktop Conversion Data
Most landing page visits now happen on phones. In Unbounce’s 2024 sample, 83% of visits were on mobile devices, and desktop visitors converted 8% better on average. The gap is not universal: mobile visitors converted better on legal pages (21% against 15.9%) and finance pages, while professional services showed a 40% desktop advantage. These mobile and desktop figures are averages and sit well above the industry medians in the first table.
Contentsquare’s whole-site data points the same way: mobile was 69.9% of traffic in the fourth quarter of 2025, but the desktop conversion rate was 74% higher. StatCounter’s page-view data shows how much the device mix varies by market. In September 2026, mobile was 59.06% of page views worldwide and 49.04% in the United States.
| Figure | What it measures | Source |
|---|---|---|
| 83% | Share of landing page visits made on mobile devices, all industries; landing pages built on Unbounce, July 2023 to July 2024 | Unbounce, Aug 29, 2024 · data Jul 2023–Jul 2024Benchmark (platform data) |
| 8% | How much higher desktop visitors’ average conversion rate was than mobile visitors’, all industries; landing pages built on Unbounce, July 2023 to July 2024 | Unbounce, Aug 29, 2024 · data Jul 2023–Jul 2024Benchmark (platform data) |
| 11.6% vs 8.3% | Professional services: average conversion rate of desktop vs mobile visitors, with 81% of visits on mobile; landing pages built on Unbounce, July 2023 to July 2024 | Unbounce, Aug 29, 2024 · data Jul 2023–Jul 2024Benchmark (platform data) |
| 21% vs 15.9% | Legal: average conversion rate of mobile vs desktop visitors, with 88% of visits on mobile; landing pages built on Unbounce, July 2023 to July 2024 | Unbounce, Aug 29, 2024 · data Jul 2023–Jul 2024Benchmark (platform data) |
| 27.8% | Finance and insurance: how much higher mobile visitors’ average conversion rate was than desktop visitors’ (11.5% vs 9%); landing pages built on Unbounce, July 2023 to July 2024 | Unbounce, Aug 29, 2024 · data Jul 2023–Jul 2024Benchmark (platform data) |
| 6.4% vs 6.2% | SaaS: average conversion rate of mobile vs desktop visitors, with 79% of visits on mobile; landing pages built on Unbounce, July 2023 to July 2024 | Unbounce, Aug 29, 2024 · data Jul 2023–Jul 2024Benchmark (platform data) |
| 22% | Health and wellness: how much higher desktop visitors’ conversion rate was than mobile visitors’, with seven times more traffic on mobile; landing pages built on Unbounce, July 2023 to July 2024 | Unbounce, Aug 29, 2024 · data Jul 2023–Jul 2024Benchmark (platform data) |
| 69.9% | Share of website traffic on mobile across 6,500+ websites in nine industries, Q4 2025 | Contentsquare, read Oct 21, 2026 · Q4 2025 dataBenchmark (platform data) |
| 74% | How much higher the desktop conversion rate was than the mobile conversion rate across 6,500+ websites, Q4 2025 (conversion rates fell on every device year on year) | Contentsquare, read Oct 21, 2026 · Q4 2025 dataBenchmark (platform data) |
| 59.06% | Mobile share of web page views worldwide, September 2026 (desktop 39.2%, tablet 1.75%); between October 2025 and September 2026 the monthly mobile share ranged from 48.99% to 59.06% | StatCounter, read Oct 21, 2026Measured (independent dataset) |
| 49.04% | Mobile share of web page views in the United States, September 2026 (desktop 48.55%, tablet 2.42%) | StatCounter, read Oct 21, 2026Measured (independent dataset) |
Copy, Readability and Attention
Simple copy converts better in most of Unbounce’s data. Pages written at a 5th to 7th grade reading level had a median conversion rate of 11.1%, against 5.3% for college-level and 5.5% for professional-level writing, and the number of difficult words correlated negatively with conversion. Unbounce notes exceptions in finance, legal and entertainment pages, where some higher reading levels converted better. The best-converting length was roughly 250 to 750 words in the SaaS and professional services data.
Attention research points the same way. Nielsen Norman Group found that users spent about 57% of their page-viewing time above the fold in 2018, down from 80% in 2010, and that users read about 20% of the words on an average page. People decide quickly whether to stay: often within 10 to 20 seconds, and a first aesthetic judgment forms in 17 to 50 milliseconds in Google researchers’ study. That is why the headline, subhead and first screen deserve most of the attention.
| Figure | What it measures | Source |
|---|---|---|
| 11.1% | Median conversion rate of landing pages written at a 5th to 7th grade reading level, 56% above 8th–9th grade copy (7.1%) and more than double college- and professional-level copy (5.3% and 5.5%); landing pages built on Unbounce, July 2023 to July 2024 | Unbounce, Aug 29, 2024 · data Jul 2023–Jul 2024Benchmark (platform data) |
| −24.3% | Correlation between the number of difficult words (three or more syllables) on a landing page and its conversion rate (−15% in the 2020 data); landing pages built on Unbounce, July 2023 to July 2024 | Unbounce, Aug 29, 2024 · data Jul 2023–Jul 2024Benchmark (platform data) |
| 12.9% vs 2.1% | SaaS: conversion rate of landing pages written at a 5th to 7th grade level vs professional-level copy; landing pages built on Unbounce, July 2023 to July 2024 | Unbounce, Aug 29, 2024 · data Jul 2023–Jul 2024Benchmark (platform data) |
| 250–725 words | SaaS: landing page length with the highest median conversion rate (3.8%); for professional services the range was 275–745 words; landing pages built on Unbounce, July 2023 to July 2024 | Unbounce, Aug 29, 2024 · data Jul 2023–Jul 2024Benchmark (platform data) |
| 57% | Share of page-viewing time spent above the fold in Nielsen Norman Group’s eyetracking study (120 participants, 130,000+ fixations, 1920×1080 screens); 74% fell in the first two screenfuls | Nielsen Norman Group, Apr 15, 2018Research (usability study) |
| 80% | Share of viewing time above the fold in Nielsen Norman Group’s 2010 eyetracking study, before long scrolling pages became common | Nielsen Norman Group, Apr 15, 2018Research (usability study) |
| 20% | Share of the words on an average web page that users read during a visit (at most 28% if they spent the whole visit reading), from browser logs of 25 users and 59,573 page views; pages averaged 593 words | Nielsen Norman Group, May 6, 2008Research (usability study) |
| 4.4 seconds | Extra time users spent on a page for each additional 100 words, enough to read about 18 of them | Nielsen Norman Group, May 6, 2008Research (usability study) |
| 10–20 seconds | Time after which users often leave a web page; the first 10 seconds are the most critical for the decision to stay or leave (analysis of Microsoft Research dwell-time data on 205,873 pages) | Nielsen Norman Group, Sep 12, 2011Research (usability study) |
| 17–50 ms | Time in which people form an aesthetic first impression of a website, in Google researchers’ study of visual complexity and prototypicality | Google Research, Aug 29, 2012Academic research |
Page Speed and Core Web Vitals
Google’s Core Web Vitals set three thresholds for a good experience, measured at the 75th percentile of real page loads: Largest Contentful Paint within 2.5 seconds, Interaction to Next Paint of 200 milliseconds or less, and Cumulative Layout Shift of 0.1 or less. Time to First Byte is a supporting metric with a rough target of 0.8 seconds.
Most sites still miss them on mobile. HTTP Archive’s Web Almanac 2025, built on Chrome user data from July 2025, found that 48% of mobile websites had good Core Web Vitals, up from 32% in 2021. Of the three metrics, loading is the weak point: 62% of mobile sites had a good LCP, and in 76% of mobile pages the largest element was an image. The median mobile home page weighed 2,362 KB, nearly three times its 2015 size.
| Figure | What it measures | Source |
|---|---|---|
| 2.5 s | Largest Contentful Paint (LCP) a page should reach for a good loading experience, measured at the 75th percentile of page loads on mobile and desktop | web.dev (Google), updated Oct 31, 2024Guideline (Google) |
| 200 ms | Interaction to Next Paint (INP) a page should stay at or below for good responsiveness, at the 75th percentile | web.dev (Google), updated Oct 31, 2024Guideline (Google) |
| 0.1 | Cumulative Layout Shift (CLS) a page should stay at or below for good visual stability, at the 75th percentile | web.dev (Google), updated Oct 31, 2024Guideline (Google) |
| 0.8 s | Time to First Byte (TTFB) most sites should aim for as a rough guide (poor above 1.8 seconds); a supporting metric, not one of the three Core Web Vitals | web.dev (Google), updated Nov 18, 2025Guideline (Google) |
| 48% | Share of mobile websites with good Core Web Vitals in Chrome user data, July 2025 (44% in 2024, 32% in 2021); desktop 56% | HTTP Archive, Jan 15, 2026Measured (independent dataset) |
| 56% vs 45% | Share of mobile secondary (inner) pages vs mobile home pages with good Core Web Vitals, July 2025 (desktop 61% vs 47%) | HTTP Archive, Jan 15, 2026Measured (independent dataset) |
| 62% | Share of mobile websites with a good Largest Contentful Paint, July 2025 (desktop 74%) | HTTP Archive, Jan 15, 2026Measured (independent dataset) |
| 77% | Share of mobile websites with a good Interaction to Next Paint, July 2025 (desktop 97%) | HTTP Archive, Jan 15, 2026Measured (independent dataset) |
| 81% | Share of mobile websites with a good Cumulative Layout Shift, July 2025 (desktop 72%) | HTTP Archive, Jan 15, 2026Measured (independent dataset) |
| 44% | Share of mobile websites with a good Time to First Byte, July 2025 (desktop 55%) | HTTP Archive, Jan 15, 2026Measured (independent dataset) |
| 76% | Share of mobile pages whose largest visible element (the LCP element) is an image, July 2025 (desktop 85.3%) | HTTP Archive, Jan 15, 2026Measured (independent dataset) |
| 16–17% | Share of pages that lazy-load their LCP image, which delays the main content, July 2025 | HTTP Archive, Jan 15, 2026Measured (independent dataset) |
| 1,916 ms | Median Total Blocking Time on mobile pages in lab tests, July 2025, up 58% from 1,209 ms in 2024 (desktop 92 ms) | HTTP Archive, Jan 15, 2026Measured (independent dataset) |
| 2,362 KB | Median weight of a mobile home page, July 2025, up from 845 KB in July 2015 | HTTP Archive, Jan 15, 2026Measured (independent dataset) |
What Speed Does to Conversions
The strongest evidence comes from controlled tests on real landing pages. Vodafone’s A/B test of a landing page with a 31% better LCP produced 8% more sales, and Rakuten 24’s one-month test of a page optimized for Core Web Vitals raised the conversion rate by 33.13%. These are single companies reported through Google’s web.dev, so treat them as examples rather than averages.
The cross-company studies are older. Deloitte’s study for Google, from 37 brand sites at the end of 2019, linked a 0.1-second improvement in mobile speed metrics with 8.4% more retail conversions and 10.1% more travel conversions. The widely repeated “7% for a delay” figure comes from Akamai’s 2017 retail report and refers to a 100-millisecond delay, not a full second. No newer cross-company study of this kind was found for this page, so these are shown with their years. For more on the revenue side, see ourpage-speed statistics and revenue impactpage.
| Figure | What it measures | Source |
|---|---|---|
| +8% | Sales from a Vodafone (Italy) landing page A/B test in which the faster version had a 31% better LCP; leads per visit rose 15% | web.dev (Google), Mar 17, 2021Case study (company-reported) |
| +33.13% | Conversion rate in a one-month Rakuten 24 landing page A/B test of a version optimized for Core Web Vitals (revenue per visitor +53.37%) | web.dev (Google), updated Aug 24, 2022Case study (company-reported) |
| +13% | Increase in conversions associated with a one-second LCP improvement on Renault’s landing pages, with a 14-point drop in bounce rate, from a correlation analysis of 10 million visits in 33 countries over four months | web.dev (Google), updated Oct 8, 2021Case study (company-reported) |
| +7% | Sales at bus-ticketing site redBus after it improved Interaction to Next Paint | web.dev (Google), updated May 10, 2023Case study (company-reported) |
| +8.4% | Retail conversions after a 0.1-second improvement in mobile site speed metrics (average order value +9.2%), across 37 European and US brand sites and over 30 million sessions, late 2019 | Deloitte, Mar 24, 2020Study (commissioned research) |
| +10.1% | Travel conversions after a 0.1-second improvement in mobile site speed metrics (average order value +1.9%), same 37-brand study, late 2019 | Deloitte, Mar 24, 2020Study (commissioned research) |
| +21.6% | Lead-generation sites: share of users progressing from the first form step to form submission after a 0.1-second speed improvement, same study | web.dev (Google), updated Jun 24, 2020Study (commissioned research) |
| 7% | Fall in conversion rates linked to a 100-millisecond delay in load time, retail websites, one month of SOASTA data (about 10 billion visits), 2017 | Akamai, Apr 19, 2017Benchmark (platform data) |
| +103% | Increase in bounce rates linked to a two-second delay in page load time, retail websites, 2017 | Akamai, Apr 19, 2017Benchmark (platform data) |
| 53% | Share of mobile site visitors who leave a page that takes longer than three seconds to load, as stated in Akamai’s 2017 retail report release | Akamai, Apr 19, 2017Benchmark (platform data) |
| 3x | Conversion rate of B2B lead-generation pages loading in 1 second vs 5 seconds (5x vs 10 seconds), 14 B2B sites, 30 days, 2022 | Portent, Apr 20, 2022Benchmark (agency research) |
| 2.5x | E-commerce conversion rate of pages loading in 1 second vs 5 seconds, six B2C e-commerce sites, 30 days, 2022 | Portent, Apr 20, 2022Benchmark (agency research) |
| +123% | Rise in the probability of a mobile visitor bouncing as load time goes from 1 to 10 seconds, from a neural-network model trained on bounce and conversion data (Google/SOASTA research, 2017) | Think with Google, Feb 2018 (first published Feb 2017) · archived May 24, 2024Model estimate (Google research) |
| −95% | Fall in the modelled probability of conversion as the number of page elements (text, titles, images) rises from 400 to 6,000, same Google model | Think with Google, Feb 2018 (first published Feb 2017) · archived May 24, 2024Model estimate (Google research) |
| ~15 s | Average time to fully load a mobile ad landing page, from Google’s test of 11 million mobile ad landing pages in 213 countries on 4G, January 2018 (70% took over 5 seconds to show above-the-fold content) | Think with Google, Feb 2018 (first published Feb 2017) · archived May 24, 2024Measured (Google research) |
Forms, Checkout Fields and Accessibility
None of the sources checked for this page gives landing page conversion rates by number of form fields, so tables claiming one should be read with care. The best-documented field counts come from checkout research. Baymard Institute found that the average e-commerce checkout in 2024 had 11.3 form fields, while its usability testing suggests seven or eight are enough, and 17% of US online shoppers say they have abandoned an order because checkout was too long or complicated.
Accessibility problems also stop people completing forms. WebAIM’s February 2026 scan of the top 1,000,000 home pages found missing form input labels on 51% of them, and a third of all form inputs were not properly labeled. Low-contrast text, which makes buttons and field hints harder to read, appeared on 83.9% of home pages.
| Figure | What it measures | Source |
|---|---|---|
| 11.3 | Form fields in the average e-commerce checkout in 2024, spread over 5.1 steps, in Baymard’s benchmark of leading sites | Baymard Institute, Jun 26, 2024Benchmark (UX audit) |
| 7–8 | Form fields an ideal checkout needs (12–14 form elements in total), according to Baymard’s usability testing | Baymard Institute, updated Sep 22, 2025Benchmark (UX audit) |
| 17% | US online shoppers who have abandoned an order because of a “too long / complicated checkout process” (excluding those who were just browsing) | Baymard Institute, updated Sep 22, 2025Survey |
| 18% | US online shoppers who have abandoned an order because the site wanted them to create an account (excluding those who were just browsing) | Baymard Institute, updated Sep 22, 2025Survey |
| 70.22% | Average documented online shopping cart abandonment rate across 50 published studies | Baymard Institute, updated Sep 22, 2025Estimate (average of published studies) |
| 51% | Share of the top 1,000,000 home pages with at least one form input missing a label, February 2026 | WebAIM, Mar 30, 2026 · February 2026 dataMeasured (independent dataset) |
| 33.1% | Share of form inputs on the top 1,000,000 home pages that were not properly labeled (6.9 inputs per home page on average), February 2026 | WebAIM, Mar 30, 2026 · February 2026 dataMeasured (independent dataset) |
| 95.9% | Share of the top 1,000,000 home pages with automatically detected WCAG 2 failures, February 2026 (94.8% in 2025), at 56.1 errors per page on average | WebAIM, Mar 30, 2026 · February 2026 dataMeasured (independent dataset) |
| 83.9% | Share of the top 1,000,000 home pages with low-contrast text below WCAG 2 AA thresholds, the most common detected error, February 2026 | WebAIM, Mar 30, 2026 · February 2026 dataMeasured (independent dataset) |
Trust and Social Proof
Reviews are a social-proof element with published university research behind them. Northwestern University’s Spiegel Research Center found that a product with five reviews was 270% more likely to be bought than one with none, with a stronger effect on higher-priced items, and that purchase likelihood peaked at ratings between 4.0 and 4.7 stars rather than at a perfect 5.0. The research is from 2017 and concerns product purchases rather than landing page conversions.
Trust also shows up at the moment of commitment: in Baymard’s survey, 19% of US online shoppers who abandoned an order said they did not trust the site with their card details.
| Figure | What it measures | Source |
|---|---|---|
| 270% | How much more likely shoppers were to buy a product with five reviews than one with none, in research by Northwestern University’s Medill Spiegel Research Center | Spiegel Research Center, 2017 research (page updated Dec 4, 2025)Academic research |
| 380% | Increase in conversion rate when reviews were displayed for a higher-priced product (190% for a lower-priced product) | Spiegel Research Center, 2017 research (page updated Dec 4, 2025)Academic research |
| 4.0–4.7 | Average star rating range at which purchase likelihood typically peaked, before falling as ratings approach 5.0 | Spiegel Research Center, 2017 research (page updated Dec 4, 2025)Academic research |
| 19% | US online shoppers who have abandoned an order because they did not trust the site with their credit card information (excluding those who were just browsing) | Baymard Institute, updated Sep 22, 2025Survey |
A/B Testing Benchmarks
Most experiments do not produce a winner. Optimizely reports that 12% of experiments on its platform produce a statistically significant improvement on the primary metric, across 173,000 experiments from more than 1,200 companies. Published success rates from large experimentation programs range from 8% in Airbnb search to 15% at Bing and 33% at Microsoft, with 10% at Booking.com, Google Ads and Netflix, as compiled by Ron Kohavi, Alex Deng and Lukas Vermeer.
The same paper shows why small tests mislead. When only one idea in ten truly works, about 22% of statistically significant wins are false positives even in a well-powered test, and more than half are at 20% power. A publicly shared landing page test that claimed a 337% lift had about 80 users per variant and 3% power to detect even a 10% change; the authors had never seen a real change anywhere near that size in tens of thousands of tests. Big wins do happen, such as the Bing headline test that lifted revenue 12%, but they are rare and need replication. For a structured way to run tests, see our guide tobuilding a CRO agent for A/B tests and landing pages.
| Figure | What it measures | Source |
|---|---|---|
| 12% | Share of experiments that produce a statistically significant improvement on the primary metric, across 173,000 experiments run on Optimizely by 1,200+ companies, 2018–2026 | Optimizely, read Oct 21, 2026Benchmark (platform data) |
| 34 | Experiments the median Optimizely customer company runs a year (the top 3% run over 500) | Optimizely, read Oct 21, 2026Benchmark (platform data) |
| 33% | Published success rate of experiments at Microsoft (ideas that truly improve the main metric), as compiled by Kohavi, Deng and Vermeer | Kohavi, Deng and Vermeer, Aug 14, 2022Academic research |
| 15% | Published success rate of experiments at Bing | Kohavi, Deng and Vermeer, Aug 14, 2022Academic research |
| 10% | Published success rate of experiments at Booking.com, Google Ads and Netflix; at that rate about 22% of statistically significant wins are false positives (p < 0.05, 80% power) | Kohavi, Deng and Vermeer, Aug 14, 2022Academic research |
| 8% | Published success rate of experiments in Airbnb search (false positive risk 26.4% at 80% power) | Kohavi, Deng and Vermeer, Aug 14, 2022Academic research |
| More than half | Share of statistically significant results that are false positives when experiments run at 20% power in organisations with a 10% or lower success rate | Kohavi, Deng and Vermeer, Aug 14, 2022Academic research |
| 3% | Statistical power to detect even a 10% relative change in a publicly shared landing page A/B test that claimed a 337% conversion lift with about 80 users per variant | Kohavi, Deng and Vermeer, Aug 14, 2022Academic research |
| +12% | Revenue lift from a single Bing ad-headline experiment in 2012, after the idea had been shelved for months as low priority | Harvard Business Review, Sep–Oct 2017Case study (company-reported) |
| 3.5x | Expected impact of experiments with four variations compared with a typical A/B test, according to Optimizely | Optimizely, read Oct 21, 2026Vendor claim |
AI Traffic and Automated Optimization
AI assistants are a new but still small traffic source. Across Contentsquare’s 6,500+ websites, AI-referred visits were 0.2% of traffic in the fourth quarter of 2025, and their conversion rate rose 55% year on year to 1.3%, the only channel whose conversion rate grew.
Landing page vendors now sell AI features that route visitors between page variants or generate copy. The published figures for them are vendor claims without a stated method, such as Unbounce’s statement that Smart Traffic users get 30% more sales and signups on average. None of the sources checked for this page measures AI-generated landing page copy against human-written pages. Measuring these tools properly depends on clean event tracking, which our guide toGoogle Analytics 4 adoptioncovers.
| Figure | What it measures | Source |
|---|---|---|
| 0.2% | Share of website traffic referred by AI platforms across 6,500+ websites, Q4 2025 (up 632% year on year) | Contentsquare, read Oct 21, 2026 · Q4 2025 dataBenchmark (platform data) |
| 1.3% | Conversion rate of AI-referred traffic, up 55% year on year and the only channel whose conversion rate grew, Q4 2024 to Q4 2025 | Contentsquare, read Oct 21, 2026 · Q4 2025 dataBenchmark (platform data) |
| 30% | More sales and signups on average for marketers who turn on Unbounce’s AI traffic-routing feature, Smart Traffic, according to Unbounce (no method or period stated) | Unbounce, read Oct 21, 2026Vendor claim |
How to Use These Benchmarks
Compare like with like. A landing page median (Unbounce), a per-click rate for search ads (WordStream) and a whole-site conversion rate (Contentsquare) measure different things, and a median and an average from the same data can differ threefold. Before using a benchmark, check whether it is a median or an average, what counts as a conversion, which traffic sources are included and when the data was collected.
The figures here are most useful for spotting large gaps: a paid search landing page far below the industry median, a mobile page that misses Core Web Vitals, a form with twice the fields it needs. The better yardstick is your own conversion rate by traffic source and device over time, tested with experiments large enough to detect the change you care about. For the wider traffic and device context, see ourwebsite statistics for 2026, and for page builds and conversion-path work, ourweb development services.
Widely Quoted Figures, Checked
Several landing page statistics circulate without a date, with the wrong measure, or credited to a report that does not exist. Each one below is checked against the publisher’s own page.
| Often quoted as | What the source says |
|---|---|
| The average landing page converts at 4.02% (Unbounce) | LinkedIn’s marketing blog (August 2025) attributes 4.02% to “a study conducted by Unbounce” without naming an edition, sample or period. Unbounce’s 2020 report gave a 3.2% median (9.7% average) and its 2024 report a 6.6% median. Unbounce has published no 2025 or 2026 edition. |
| The average landing page conversion rate is 2.35%, and the top quarter converts at 11.45% | Both come from a March 2014 WordStream analysis of Google Ads accounts: 2.35% was the median account, the top 25% converted at 5.31% or more and the top 10% at 11.45%. The post’s introduction calls these landing page rates, but they are account-wide conversion rates, and WordStream notes that the 11.45% figure is not for individual landing pages. Its 2026 median for US search campaigns is 8.18%. |
| The average conversion rate across industries is 9.7% | This is the average in Unbounce’s 2020 report (data from October 2018 to October 2019). The median of the same pages was 3.2%. A few very high-converting pages lift the average far above the typical page. |
| A one-second delay costs 7% of conversions | Akamai’s April 2017 release (SOASTA data, about 10 billion retail visits) says a 100-millisecond delay “can hurt conversion rates by 7 percent”. It is a 2017 retail figure for a much shorter delay, not a current landing page benchmark. |
| 53% of mobile visitors leave a page that takes longer than three seconds to load | Akamai’s April 2017 retail report release lists this among its findings. It is a 2017 figure. For a current measure, HTTP Archive found that 62% of mobile websites had a good Largest Contentful Paint (2.5 seconds or less) in July 2025. |
| Bounce probability rises 123% as load time goes from one to ten seconds | Google’s February 2018 Think with Google article gives this as the output of a neural-network model trained on bounce and conversion data (Google/SOASTA research, 2017). It is a model estimate for mobile sites, and the article has since been removed from Think with Google. |
| Pages with video convert 86% better | This traces to a sales brochure from the video-ad company EyeView describing a single A/B test on one tutoring company’s landing page, which reported “an 86 percent increase in Conversion”. It is one vendor case study, not a study of landing pages in general. |
| 57% of visitors never scroll below the fold | Nielsen Norman Group’s 2018 eyetracking study found that users spent about 57% of their page-viewing time above the fold, and 74% in the first two screenfuls. It measures where viewing time goes, not how many visitors scroll. |
| Only 13% of A/B tests produce a significant winner | None of the sources checked for this page gives 13%. Optimizely reports 12% across 173,000 experiments, and published rates at Microsoft, Bing, Booking.com, Google Ads, Netflix and Airbnb range from 8% to 33%. |
| Three-field forms convert at 10.1% and nine-field forms at 3.6% (Unbounce) | Unbounce’s 2024 report does not publish conversion rates by number of form fields. The best-documented field counts come from Baymard’s checkout research: 11.3 fields in the average 2024 checkout against seven or eight in an ideal one. |
Quote the Sample, the Measure and the Year
A landing page benchmark is only useful with its context. “The median landing page built on Unbounce converted at 6.6% between July 2023 and July 2024” can be checked and compared; “the average landing page converts at 4.02%” cannot, because it names no sample, measure or year.
Use published figures to find large gaps, then judge your pages against your own conversion rate by traffic source and device, with tests big enough to tell a real change from noise.
Sources
Every figure on this page comes from one of the sources below.
- HTTP Archive, Web Almanac 2025Performance chapter (CrUX and HTTP Archive data, July 2025)Jan 15, 2026
- HTTP Archive, Web Almanac 2025Page Weight chapter (July 2025 crawl)Jan 15, 2026
- StatCounter Global StatsDesktop vs Mobile vs Tablet Market Share Worldwide (monthly, September 2026)read Oct 21, 2026
- StatCounter Global StatsDesktop vs Mobile vs Tablet Market Share United States Of America (monthly, September 2026)read Oct 21, 2026
- WebAIM, Institute for Disability Research, Policy, and Practice, Utah State UniversityThe WebAIM Million: The 2026 report on the accessibility of the top 1,000,000 home pagesMar 30, 2026 · February 2026 data
- web.dev (Google)Web Vitalsupdated Oct 31, 2024
- web.dev (Google)Time to First Byte (TTFB)updated Nov 18, 2025
- web.dev (Google)Vodafone: A 31% improvement in LCP increased sales by 8%Mar 17, 2021
- web.dev (Google)How Rakuten 24’s investment in Core Web Vitals increased revenue per visitor by 53.37% and conversion rate by 33.13%updated Aug 24, 2022
- web.dev (Google)How Renault improved its bounce and conversion rates by measuring and optimizing Largest Contentful Paintupdated Oct 8, 2021
- web.dev (Google)How redBus improved their website’s Interaction to Next Paint (INP) and increased sales by 7%updated May 10, 2023
- web.dev (Google)Milliseconds make millions (study summary)updated Jun 24, 2020
- Think with Google (Google)Find out how you stack up to new industry benchmarks for mobile page speed (Daniel An; archived copy, the original now redirects)Feb 2018 (first published Feb 2017) · archived May 24, 2024
- Google ResearchUsers love simple and familiar designs – Why websites need to make a great first impressionAug 29, 2012
- Deloitte Ireland, commissioned by Google, with data from Fifty-FiveMilliseconds Make Millions: A study on how improvements in mobile site speed positively affect a brand’s bottom lineMar 24, 2020
- Nielsen Norman GroupScrolling and AttentionApr 15, 2018
- Nielsen Norman GroupHow Little Do Users Read?May 6, 2008
- Nielsen Norman GroupHow Long Do Users Stay on Web Pages?Sep 12, 2011
- Ron Kohavi, Alex Deng and Lukas Vermeer (KDD ’22, ACM)A/B Testing Intuition Busters: Common Misunderstandings in Online Controlled ExperimentsAug 14, 2022
- Harvard Business Review (Ron Kohavi and Stefan Thomke)The Surprising Power of Online ExperimentsSep–Oct 2017
- Medill Spiegel Research Center, Northwestern UniversityHow Online Reviews Influence Sales2017 research (page updated Dec 4, 2025)
- Akamai Technologies (investor relations release)Akamai Online Retail Performance Report: Milliseconds Are CriticalApr 19, 2017
- UnbounceConversion Benchmark Report 2024 (overview, methodology and nine industry pages)Aug 29, 2024 · data Jul 2023–Jul 2024
- Unbounce2020 Conversion Benchmark Report (and methodology page)Apr 17, 2020 · data Oct 2018–Oct 2019
- UnbounceSmart Traffic product pageread Oct 21, 2026
- WordStream by LocaliQGoogle Ads Benchmarks 2026: Competitive Data & Insights for Every IndustryMay 18, 2026 · Apr 2025–Mar 2026 data
- WordStream by LocaliQFacebook Ads Benchmarks 2026: NEW Data, Trends, & Insights for Your IndustrySep 14, 2026 · Apr 2025–Jun 2026 data
- WordStreamWhat’s a Good Conversion Rate? (It’s Higher Than You Think)Mar 17, 2014
- Contentsquare2026 Digital Experience Benchmark (report guide and conversion benchmarks)read Oct 21, 2026 · Q4 2025 data
- OptimizelyTop 10 takeaways from running 173,000 experiments, and 173K experiments later (2026 report; data: 173k experiments, 1,200+ companies, 2018–2026)read Oct 21, 2026
- PortentSite Speed is (Still) Impacting Your Conversion RateApr 20, 2022
- LinkedIn Marketing BlogHow to Use LinkedIn Lead Gen FormsAug 19, 2025
- EyeView (video advertising company)Sales brochure with eToro and TutorVista video case studiesundated (copy online since at least 2018)
- Baymard InstituteCheckout Optimization: 5 Ways to Minimize Form Fields in CheckoutJun 26, 2024
- Baymard Institute50 Cart Abandonment Rate Statistics 2026updated Sep 22, 2025
Turn Landing Page Data Into Better Pages
We audit, rebuild and test landing pages against your own conversion data, traffic mix and page speed, not generic averages.
Deep dives on AI, marketing and development.
Practical guides and fresh insights by email. No recycled takes.
Related Articles
Continue exploring with these related guides