2025: 0 of the Global Top 200 Websites Use Valid HTML

Published on Sep 10, 2025, filed under development, html, conformance (feed). (Share this on Mastodon or Bluesky?)

It’s time for the annual check on HTML conformance, reviewing whether the HTML code in our field is error-free and valid. The short version: None of the most-frequented sites on this planet uses valid HTML.

The usual disclaimer: The following data gives an impression of precision that is greater than warranted. The point of this annual analysis is to check on full HTML conformance—absence of markup errors—on home pages. That is, the specific error counts don’t matter for the purpose of telling conformance, and pages other than home pages aren’t being checked. I’m providing the data for further study and for comparability to previous years.

Contents

1. Analysis
2. HTML Conformance in 2025
3. HTML Conformance Over Time
4. Notes and Observations
5. Interpretation

Analysis

Like every year since 2021 (cf. 2022, 2023, 2024) I used the the Ahrefs Top 1,000 most visited websites to check the home pages of the first 200 websites on HTML conformance. ^*

For this purpose, I took all the respective URLs, prepared HTML validation URLs, validated each page, and documented the error counts of the 200 tests in a spreadsheet:

HTML Conformance in 2025

Degraded: 0 of the 200 home pages have no HTML conformance errors. This is down 100% from the 1 home page that was valid in 2024.

Improved: 6 home pages (up from 5 in 2024) have only 1 validation error.

Degraded: 37 additional home pages (down from 44) have a single-digit number of validation errors.

Degraded: 63 home pages (up from 56) have more than 100 errors and may or may not resemble HTML.

Degraded: The average number of HTML errors is 110.45 (up from 99.34).

Degraded: The median is 55.5 (up from 35.5 last year).

HTML Conformance Over Time

We need to test more websites for this to be significant, but here’s how error counts developed over the years:

Notes and Observations

The W3C HTML validator (i.e., the “Nu” portion handling living HTML) still makes work on analyzing (and fixing) HTML unnecessarily difficult, because it blends HTML with CSS and other errors. This continues to slow down the work on these analyses, and introduces a potential error source by requiring to manually deduct non-HTML errors from the error counts. (If you like to follow and add your voice to the respective thread, note Nu Validator issue #940.)

The W3C validator also undercounts, because it validates static HTML. It may also validate interstitials instead of home pages. Where I caught this, I manually validated the hydrated markup of the affected home page (blue highlighting). In general, however, the error count is likely higher, and the conformance situation in the field worse.

I didn’t pay much attention to whether pages that couldn’t be tested by the validator were actively blocking it. This used to be a problem, but this year, I was quick pivoting to either validate manually, or include another website in the sample.

As mentioned in the intro, the analysis checks on whether the most popular websites use valid HTML. For that purpose, it’s interesting but not decisive how many issues there are, or if a page to be tested doesn’t happen to be the main page. The logic is this: If any page on a website, like the home page, has validation errors, the site cannot be valid anymore. ^†

Interpretation

First, what do you think? Please share your observations as a response on Mastodon, Bluesky, or LinkedIn!

Personally, it’s still puzzling to be in a field that seems to care so little about their craft, and to regularly run into cases that demonstrate why this is neither good for any individual nor the field itself (especially not with AI’s breath behind everyone).

100% of the most popular websites in the world contain fantasy HTML. This must sound funny to an outsider, but for us as web professionals, I’m not sure there could be any bigger sign of professional insolvency. As mentioned in the past, a number of these sites are being maintained by some of the best-paid of our peers. They might be best-paid but not best, you might wonder, but we don’t know that. Yet there seems to be merit in arguing that any web developer in charge of a high-profile website should be able to check and ensure that the HTML they publish is actual, error-free HTML. That’s not difficult. That’s not a lot of work. That’s—professionalism.

If you want to know why and how to write valid HTML, check out my link guide on minimal and valid HTML. Also, stay in touch, because on this website and in my books, I write a lot about the craft of HTML. See you around.

^* I do so faithfully, that is, I don’t check and question Ahrefs’ methodology. There have been concerns about some of these sites in the past—including that they seemed spammy—, but I’m leaving that issue to Ahrefs. This is not to say that you shouldn’t be more critical.

^† Of course, this doesn’t work the other way around: If a home page has no errors, it doesn’t mean the entire website is valid—other pages could still contain errors.

About Me

I’m Jens (long: Jens Oliver Meiert), and I’m a web developer, manager, and author. I’ve been working as a technical lead and engineering manager for companies you’ve never heard of and companies you use every day, I’m an occasional contributor to web standards (like HTML, CSS, WCAG), and I write and review books for O’Reilly and Frontend Dogma.

I love trying things, not only in web development and engineering management, but also in other areas like philosophy. Here on meiert.com I share some of my experiences and views. (I value you being critical, interpreting charitably, and giving feedback.)