Data Study · Technical SEO

    ETag and Last-Modified Usage: 63% of Top 1M Sites Send Neither

    493,049 of 782,570 top-1M homepages (63.0%) send neither an ETag nor a Last-Modified header, so crawlers can't make cheap conditional requests.

    • 782,570 top-1M homepages
    • HTTP Archive, Sept 2026
    63.0%
    In this study

    Homepage validator headers by platform

    PlatformNeither headerETagLast-ModifiedHomepages
    Squarespace3.7%95.1%1.4%513
    Wix5.5%93.5%1.0%899
    Shopify12.6%86.8%0.7%47,127
    Webflow18.3%5.5%80.4%2,589
    WordPress61.7%9.7%31.2%113,580
    Other or custom67.4%19.2%26.1%617,837
    All homepages63.0%22.0%25.5%782,570

    Source: PageSpeed Matters, Sept 2026. Homepages that send both headers count in both the ETag and Last-Modified columns.

    Summary

    ETag and Last-Modified are validator headers. When a server sends one, a browser or crawler can come back later and ask "has this changed?" If it has not, the server can answer with a short 304 Not Modified instead of sending the whole page again. That saves bandwidth for the server and makes repeat fetches cheaper for crawlers.

    We checked the main HTML response of 782,570 homepages in the top 1M sites by Chrome traffic. 493,049 of them, 63.0%, send neither header. Hosted site builders almost always send one, while WordPress and custom stacks usually do not.

    What we measured

    We used the HTTP Archive crawl from September 2026, mobile, covering the top 1M sites by Chrome traffic, and read the response headers on each homepage's main document. A homepage counts as sending a header if the ETag or Last-Modified header is present at all. We did not test whether the server then answers conditional requests correctly.

    Platform is the detected site platform. Popularity tiers come from the Chrome traffic rank. Speed figures are real-user time to first byte (TTFB) from Chrome UX Report data, for homepages that have it.

    One limit matters for reading the results. Dynamic pages, such as pages that change for every visitor, may leave these headers out on purpose, so a missing validator is not always an error.

    Most homepages send no validator header

    63.0% of the 782,570 homepages we checked, 493,049 in total, send neither an ETag nor a Last-Modified header. For those pages, a crawler has no cheap way to ask whether anything changed. It has to download the full HTML again each time.

    Last-Modified is slightly more common than ETag. 25.5% of homepages (199,356) send Last-Modified and 22.0% (172,270) send an ETag. Only 10.5% (82,105) send both, so most sites that use validators pick one header and stop there.

    63.0%

    Share of 782,570 top-1M homepages that send neither an ETag nor a Last-Modified header.

    Hosted builders almost always send a validator

    The platform a site runs on is the clearest pattern in the data. Hosted builders almost always send a validator: only 3.7% of 513 Squarespace homepages, 5.5% of 899 Wix homepages and 12.6% of 47,127 Shopify homepages send neither header.

    Self-managed stacks are different. 61.7% of 113,580 WordPress homepages send neither header, and so do 67.4% of 617,837 homepages on other or custom stacks. On hosted platforms the provider configures the server and CDN once for every site. On WordPress and custom builds, the result depends on each site's hosting, caching plugin and server setup, and many of those setups leave validators out.

    Platforms also tend to pick one header. Shopify (86.8%), Wix (93.5%) and Squarespace (95.1%) homepages mostly send an ETag, while Webflow homepages mostly send Last-Modified: 80.4% of 2,589 do. Webflow sits between the two groups, with 18.3% sending neither.

    Popularity makes almost no difference

    Bigger sites are no better at this. 64.6% of the 754 Top 1K homepages send neither header, against 62.9% of the 707,768 homepages ranked below 100,000. The most visited sites on the web skip validators about as often as everyone else.

    Traffic, in other words, has little to do with whether a site sends validators. Platform and server setup line up with it far more closely, as the gap between hosted builders and self-managed stacks shows.

    Homepages without validators have slower TTFB

    Among homepages with real-user data, those that send neither header have a median TTFB of 793 ms. Homepages that send at least one have a median of 645 ms.

    This is a correlation, not a cause. A validator header does not make the first response faster on its own. The more likely link is the setup behind it: pages that are served from a cache or a static build usually carry a validator and come back quickly, while pages generated fresh on every request often skip validators and take longer to produce.

    What this means for your site

    Start by checking what your homepage sends. Run a header check on the URL, or open your browser's network panel, select the main document and look for ETag or Last-Modified in the response headers.

    • If the page is static or cached, turn on ETag or Last-Modified at the server or CDN. Most web servers and CDNs support both.
    • Check that conditional requests actually work. Send the request again with If-None-Match or If-Modified-Since and confirm the server answers 304 Not Modified when nothing has changed.
    • Make ETags stable. An ETag that changes on every request, or differs between servers behind a load balancer, gives crawlers nothing to compare.
    • On WordPress, a page caching plugin or host-level cache is usually where these headers get set, so check its settings first.
    • For pages that really are different for each visitor, leaving validators out can be the right call.

    Sources

    • HTTP Archive crawl, September 2026, mobile, top 1M sites by Chrome traffic, main document response headers
    • Chrome UX Report (CrUX) real-user TTFB, September 2026
    • MDN Web Docs, HTTP caching: https://developer.mozilla.org/en-US/docs/Web/HTTP/Caching

    Embed this research

    Paste this on your site to embed the chart. It links back to the source automatically.

    <iframe id="psm-etag-and-last-modified-usage" src="https://www.pagespeedmatters.com/embed/data-studies/etag-and-last-modified-usage" width="100%" height="600" style="width:100%;border:1px solid #E2E3E1;border-radius:12px" loading="lazy" title="ETag and Last-Modified Usage: 63% of Top 1M Sites Send Neither - PageSpeed Matters"></iframe>
    <script>window.addEventListener("message",function(e){if(e&&e.data&&e.data.psmEmbed==="etag-and-last-modified-usage"&&e.data.height){var f=document.getElementById("psm-etag-and-last-modified-usage");if(f){f.style.height=e.data.height+"px";}}});</script>
    <p style="font:14px/1.5 system-ui,sans-serif">Source: <a href="https://www.pagespeedmatters.com/resources/data-studies/etag-and-last-modified-usage">ETag and Last-Modified Usage: 63% of Top 1M Sites Send Neither - PageSpeed Matters</a></p>

    Frequently asked questions

    What percentage of websites use ETag or Last-Modified headers?

    Across 782,570 top-1M homepages, 22.0% send an ETag, 25.5% send Last-Modified and 10.5% send both. 63.0% send neither header.

    Which platforms send ETag headers?

    Hosted builders mostly send an ETag: 95.1% of Squarespace, 93.5% of Wix and 86.8% of Shopify homepages do. Webflow homepages mostly send Last-Modified instead (80.4% of 2,589).

    Do WordPress sites send ETag or Last-Modified headers?

    Most do not. 61.7% of 113,580 WordPress homepages send neither header, against 12.6% of Shopify homepages and 3.7% of Squarespace homepages.

    How do I check if my site sends ETag or Last-Modified?

    Open your browser's network panel, select the main document and look for ETag or Last-Modified in the response headers, or use a header checker. Then repeat the request with If-None-Match or If-Modified-Since and confirm the server returns 304 Not Modified.

    Validator headers are usually a server or CDN setting, and turning them on lets crawlers check for changes without downloading every page again. Need help with this? See our technical SEO services.

    Share

    How we measured

    Methodology & Source

    Data source
    HTTP Archive crawl of the top 1M sites (mobile, Sept 2026) with Chrome UX Report TTFB field data
    Collection date
    September 2026
    Sample
    782,570 top-1M homepages
    Release
    HTTP Archive, Sept 2026

    Homepage main document response headers from the HTTP Archive Sept 2026 mobile crawl of the top 1M sites by Chrome traffic; 782,570 homepages. A header counts when it is present. Dynamic pages may omit these headers on purpose, so not every omission is an error. The TTFB gap is a correlation, not a cause.

    Using this data?

    Cite this study

    Link back to the study when you cite it.

    APA

    Matt Suffoletto. (2026). ETag and Last-Modified Usage: 63% of Top 1M Sites Send Neither. PageSpeed Matters. https://www.pagespeedmatters.com/resources/data-studies/etag-and-last-modified-usage

    Plain link

    ETag and Last-Modified Usage: 63% of Top 1M Sites Send Neither - PageSpeed Matters - https://www.pagespeedmatters.com/resources/data-studies/etag-and-last-modified-usage

    Ready to Get Started?

    Get your site optimized in as little as 3 days. From $500.