Web design statistics at a glance
Web design in 2024 is a story of scale, friction, and missed basics. The dataset behind this article shows that many home pages still struggle with accessibility fundamentals, while speed and checkout experience remain major conversion risks across the broader web.
Fast facts
- 81% of home pages had low-contrast text below WCAG 2 AA thresholds (WebAIM Million 2024).
- Home pages averaged 34.5 distinct low-contrast text instances each (WebAIM Million 2024).
- 21.6% of all home page images were missing alternative text (WebAIM Million 2024).
- 35.5% of form inputs were not properly labeled (WebAIM Million 2024).
- 37.9% of pages had skipped heading levels (WebAIM Million 2024).
- 64% of desktop sites had mediocre or worse checkout UX performance (Baymard 2025 Checkout UX).
- Google guidance in the dataset says above-the-fold content should arrive in under 1 second on mobile networks (Google PageSpeed Insights mobile guidance).
What this article covers
- The most important web design statistics in the dataset.
- Accessibility patterns across contrast, alt text, headings, landmarks, and ARIA.
- Speed and checkout data that shape the business impact of design decisions.
- Benchmarks and before/after examples that show what improvement can look like.
Web design accessibility statistics
Accessibility is the clearest theme in the dataset. The numbers point to a web where visual clarity, semantic structure, and assistive technology support still break down at scale.
Big number: 81% of home pages had low-contrast text below WCAG 2 AA thresholds (WebAIM Million 2024). That is not a small edge case. It means most home pages in the sample included at least one contrast problem, and the average page had 34.5 distinct low-contrast text instances (WebAIM Million 2024).
That second figure matters because it suggests contrast problems are often repeated within the same page, not isolated to a single element. A design system with weak token discipline can replicate the same accessibility failure across buttons, nav items, labels, and supporting text.
Accessibility callouts
- 21.6% of all home page images had missing alternative text (WebAIM Million 2024).
- Home pages averaged 12 images with missing alternative text per page (WebAIM Million 2024).
- 43% of images missing alternative text were linked images (WebAIM Million 2024).
- One out of every four linked images was missing alternative text (WebAIM Million 2024).
- 14.6% of images with alternative text used questionable or repetitive alt text (WebAIM Million 2024).
The pattern is important: missing alt text is only part of the issue. The dataset also shows that when alt text exists, quality still varies enough to create repetitive or low-value descriptions. Good web design is not only about adding attributes; it is about writing meaningful, context-aware labels.
Accessibility by structure
Structural issues show up just as clearly as visual ones.
- 23.8 million headings were detected across the million pages (WebAIM Million 2024).
- Home pages averaged nearly 24 headings each (WebAIM Million 2024).
- 1.5 million h1 headings were detected, or 6.4% of all headings (WebAIM Million 2024).
- 7.3 million h2 headings were detected, or 30.5% of all headings (WebAIM Million 2024).
- 8.6 million h3 headings were detected, or 36.3% of all headings (WebAIM Million 2024).
- 3.7 million h4 headings were detected, or 15.6% of all headings (WebAIM Million 2024).
- 1.8 million h5 headings were detected, or 7.5% of all headings (WebAIM Million 2024).
- 0.9 million h6 headings were detected, or 3.7% of all headings (WebAIM Million 2024).
- Total headings decreased 4% from 24.8 million in 2023 (WebAIM Million 2024).
A concise comparison helps make sense of the heading mix:
| Metric | 2024 statistic | Source label |
|---|---|---|
| h1 headings | 1.5 million | WebAIM Million 2024 |
| h2 headings | 7.3 million | WebAIM Million 2024 |
| h3 headings | 8.6 million | WebAIM Million 2024 |
| h4 headings | 3.7 million | WebAIM Million 2024 |
| h5 headings | 1.8 million | WebAIM Million 2024 |
| h6 headings | 0.9 million | WebAIM Million 2024 |
| Total headings vs. 2023 | down 4% | WebAIM Million 2024 |
The distribution leans heavily toward h2 and h3. That is often what you want in a content-heavy layout, but the dataset also shows that structure is not always used cleanly: 16.8% of home pages had more than one h1 element, down from 20.1% in 2023 (WebAIM Million 2024). Meanwhile, 37.9% of pages had skipped heading levels and 11.3% had no headings at all (WebAIM Million 2024).
Those numbers point to two opposite failures. Some pages overuse the top-level heading, while others leave the hierarchy incomplete or absent. Both make it harder for users and assistive technologies to scan the page.
Landmark and ARIA usage
The dataset shows widespread use of ARIA and landmark semantics, but not always with cleaner outcomes.
- 76.7% of home pages had at least one region or ARIA landmark (WebAIM Million 2024).
- 39.1% had a main landmark or
<main>element (WebAIM Million 2024). - 15.5% had a search landmark (WebAIM Million 2024).
- 89,050,856 ARIA attributes were detected (WebAIM Million 2024).
- Pages averaged 89 ARIA attributes each (WebAIM Million 2024).
- ARIA code usage increased 15% in one year and has more than quadrupled since 2019 (WebAIM Million 2024).
- Pages averaged 19.2 aria-label, aria-labelledby, or aria-describedby attributes (WebAIM Million 2024).
- 74.6% of home pages used ARIA outside landmark roles (WebAIM Million 2024).
- Pages with ARIA present had 34.2% more detected errors than pages without ARIA (WebAIM Million 2024).
- Those pages had about 15 additional potential barriers on average (WebAIM Million 2024).
Why it matters
ARIA is powerful, but the dataset suggests it is often used as a patch rather than as part of a clean semantic foundation. That is reflected in the fact that pages with ARIA present had more detected errors on average (WebAIM Million 2024). The design lesson is straightforward: use native HTML first, then add ARIA where it actually improves accessibility.
Web design performance and speed statistics
Design does not stop at appearance. Page load speed shapes whether users ever experience the interface you designed.
The speed guidance in the dataset is blunt. Google recommended delivering above-the-fold content in under 1 second on mobile networks (Google PageSpeed Insights mobile guidance). It also said delays longer than 1 second can interrupt the user’s flow of thought, and it recommended keeping above-the-fold content under 98 KB to allow painting after three roundtrips (Google PageSpeed Insights mobile guidance). Google also recommended reserving 200 ms for browser layout and rendering overhead (Google PageSpeed Insights mobile guidance).
Mobile speed benchmarks
- 53% of visits are likely to be abandoned if pages take longer than 3 seconds to load (Google The Need for Mobile Speed).
- One out of two people expected a page to load in less than 2 seconds (Google The Need for Mobile Speed).
- 46% of people said waiting for pages to load was what they disliked most about browsing on mobile (Google The Need for Mobile Speed).
- The average load time for mobile sites was 19 seconds over 3G connections (Google The Need for Mobile Speed).
- The average size of content on mobile sites was 1.49 MB (Google The Need for Mobile Speed).
- That average took 7 seconds to load over 3G (Google The Need for Mobile Speed).
- Mobile pages made an average of 214 server requests (Google The Need for Mobile Speed).
- Nearly half of all server requests were ad-related (Google The Need for Mobile Speed).
These are not purely technical metrics. They describe the gap between a design mockup and a real user experience. A beautiful page that loads slowly is still a poor page.
What faster design can unlock
The dataset includes a useful before/after example from The Economic Times (web.dev Economic Times case study):
- Over 45 million monthly active users were served by The Economic Times (web.dev Economic Times case study).
- Overall CLS improved by 250%, from 0.25 to 0.09 (web.dev Economic Times case study).
- Overall LCP improved by 80%, from 4.5 seconds to 2.5 seconds (web.dev Economic Times case study).
- Poor LCP values fell from 15.03% to 10.08% (web.dev Economic Times case study).
- Poor CLS values fell from 25.92% to 9% (web.dev Economic Times case study).
- Good CLS values rose from 62.62% to 76.5% (web.dev Economic Times case study).
- Bounce rate fell by 43% overall (web.dev Economic Times case study).
- FID dropped from 200 ms to under 50 ms (web.dev Economic Times case study).
- The site achieved a 75th-percentile LCP of 2.5 seconds after optimization (web.dev Economic Times case study).
- The site achieved a 75th-percentile CLS of 0.09 after optimization (web.dev Economic Times case study).
Speed and business impact table
| Metric | Before | After | Source label |
|---|---|---|---|
| CLS | 0.25 | 0.09 | web.dev Economic Times case study |
| LCP | 4.5 seconds | 2.5 seconds | web.dev Economic Times case study |
| FID | 200 ms | under 50 ms | web.dev Economic Times case study |
| Poor LCP share | 15.03% | 10.08% | web.dev Economic Times case study |
| Poor CLS share | 25.92% | 9% | web.dev Economic Times case study |
| Good CLS share | 62.62% | 76.5% | web.dev Economic Times case study |
| Bounce rate | baseline | down 43% | web.dev Economic Times case study |
The table shows why speed work belongs in any serious web design plan. Performance changes did not just make the page faster; they improved core user experience outcomes and reduced bounce.
Web design checkout statistics
For ecommerce, checkout is where design choices become revenue outcomes. The dataset includes both broad performance benchmarks and specific operational clues about checkout complexity.
Checkout UX benchmarks
- 64% of desktop sites had mediocre or worse Checkout UX performance (Baymard 2025 Checkout UX).
- 63% of mobile sites had mediocre or worse Checkout UX performance (Baymard 2025 Checkout UX).
- 46% of apps had mediocre or worse Checkout UX performance (Baymard 2025 Checkout UX).
- 70% of ecommerce users abandon their purchase after adding items to the cart (Baymard 2025 Checkout UX).
- Large ecommerce sites can gain as much as a 35% conversion-rate increase by improving checkout design (Baymard 2025 Checkout UX).
- Baymard’s 2025 checkout benchmark includes 41,000+ manually reviewed checkout performance scores (Baymard 2025 Checkout UX).
- It also includes 33,000+ worst and best practice examples (Baymard 2025 Checkout UX).
Checkout fields and friction
The form-field data makes the friction more concrete:
- The average checkout flow in 2024 was 5.1 steps long (Baymard 2024 Form Fields).
- The average checkout contained 11.3 form fields (Baymard 2024 Form Fields).
- 18% of users had abandoned due to checkout complexity (Baymard 2024 Form Fields).
- Checkout usability starts to suffer after 8+ steps (Baymard 2024 Form Fields).
These figures give designers a practical benchmark. The question is not whether a checkout looks polished. It is whether the flow stays short enough, clear enough, and low-friction enough to survive the user’s intent.
What the checkout numbers suggest
- Desktop, mobile, and app checkout experiences all show widespread room for improvement (Baymard 2025 Checkout UX).
- Complexity is still a meaningful abandonment driver, not just a minor annoyance (Baymard 2024 Form Fields).
- Every extra field and step has to justify itself.
- Clear design decisions can be worth measurable conversion gains (Baymard 2025 Checkout UX).
Web design statistics by priority area
The strongest way to use this dataset is to group the findings into priority buckets. That turns a long list of numbers into a design action plan.
1. Accessibility basics
- Fix low-contrast text first, because the problem appears on 81% of home pages (WebAIM Million 2024).
- Audit image alt text, because 21.6% of home page images are missing it (WebAIM Million 2024).
- Check heading hierarchy, because 37.9% of pages have skipped heading levels (WebAIM Million 2024).
- Review form labels, because 35.5% of form inputs are not properly labeled (WebAIM Million 2024).
2. Semantics and structure
- Reduce redundant or broken heading structures.
- Use native landmarks consistently, because only 39.1% of home pages had a main landmark or
<main>element (WebAIM Million 2024). - Treat ARIA as a precision tool, not a substitute for semantic HTML (WebAIM Million 2024).
3. Speed and content weight
- Keep above-the-fold content lean enough to meet the under 1 second guidance for mobile networks (Google PageSpeed Insights mobile guidance).
- Watch the size and request count of mobile pages, since the dataset reports 1.49 MB of content and 214 server requests on average (Google The Need for Mobile Speed).
- Cut ad overhead where possible, because nearly half of requests were ad-related (Google The Need for Mobile Speed).
4. Checkout design
- Remove needless fields and steps.
- Protect users from confusion, because checkout complexity caused abandonment for 18% of users (Baymard 2024 Form Fields).
- Benchmark against real-world checkout patterns, not internal preference alone (Baymard 2025 Checkout UX).
Key patterns in the web design data
A few themes appear again and again across the dataset.
- Accessibility is still the most obvious baseline failure. Contrast, alt text, headings, labels, and landmarks all show measurable weakness (WebAIM Million 2024).
- Speed remains a design problem, not just an engineering problem. The mobile data shows how load time, payload size, and request volume translate directly into lost attention (Google The Need for Mobile Speed).
- Checkout design is an economic multiplier. Small UX changes can influence conversion, abandonment, and bounce at scale (Baymard 2025 Checkout UX; Baymard 2024 Form Fields).
- Better metrics often travel together. The Economic Times example shows that improving LCP, CLS, and FID can reduce bounce and improve real-world behavior (web.dev Economic Times case study).
Practical reading of the numbers
If you are designing or redesigning a site, this dataset points to a simple order of operations.
- Start with visual accessibility.
- Check the semantic skeleton next.
- Then reduce performance drag.
- Finally, audit high-friction flows like checkout and forms.
That order makes sense because the dataset shows foundational issues at every layer. A page can have strong branding and still fail contrast checks, load too slowly, or create avoidable checkout friction.
Web design statistics summary table
| Topic | Notable statistic | Source label |
|---|---|---|
| Contrast | 81% of home pages had low-contrast text | WebAIM Million 2024 |
| Images | 21.6% of home page images missed alt text | WebAIM Million 2024 |
| Headings | 37.9% of pages had skipped heading levels | WebAIM Million 2024 |
| Landmarks | 39.1% had a main landmark or <main> | WebAIM Million 2024 |
| ARIA | 89,050,856 ARIA attributes detected | WebAIM Million 2024 |
| Speed | 53% of visits may be abandoned after 3 seconds | Google The Need for Mobile Speed |
| Mobile UX | Average mobile site load time was 19 seconds over 3G | Google The Need for Mobile Speed |
| Checkout | 70% of ecommerce users abandon after cart addition | Baymard 2025 Checkout UX |
| Checkout flow | Average checkout was 5.1 steps and 11.3 fields | Baymard 2024 Form Fields |
| Optimization | The Economic Times reduced bounce by 43% overall | web.dev Economic Times case study |
Web design statistics worth remembering
- 81% low-contrast home pages (WebAIM Million 2024).
- 21.6% of home page images missing alt text (WebAIM Million 2024).
- 37.9% of pages with skipped heading levels (WebAIM Million 2024).
- 53% likely abandonment rate after 3 seconds of load time (Google The Need for Mobile Speed).
- 70% cart abandonment after adding items (Baymard 2025 Checkout UX).
- 43% bounce reduction in the Economic Times case study after performance improvements (web.dev Economic Times case study).