Is it true that Googlebot really favors HTTPS for crawling your site?

What does this 75/25 split really mean for crawling?

Google states that 75% of its crawl volume is now done through HTTPS. This isn't a new announcement — the trend has been visible for several years — but it's the first time Google has communicated such a precise ratio.

However, this global statistic masks a heterogeneous reality. Sites that have not yet migrated to HTTPS — which still represent a significant portion of the web — continue to be crawled in HTTP. The 25% of residual HTTP traffic reflects these sites, as well as URLs with errors, misconfigured redirects, or pages indexed in HTTP that haven't yet been replaced in the index by their HTTPS version.

Does Google treat HTTP and HTTPS sites differently during crawling?

The short answer: yes, but indirectly. Google has not explicitly penalized HTTP sites in its ranking algorithm since 2014, but HTTPS is considered a light trust signal. What really changes is how Googlebot allocates its resources.

A site in HTTPS benefits from a better crawl efficiency: HTTP/2 (which requires HTTPS) allows multiplexing, meaning multiple simultaneous requests over a single connection. As a result, Googlebot can explore more pages in less time. Given equal crawl budget, a HTTPS site will theoretically be crawled more effectively than a comparable HTTP site — especially at very large volumes.

Does this evolution change anything for sites already in HTTPS?

If your site is already 100% HTTPS with a clean setup — valid certificate, 301 HTTP→HTTPS redirects, consistent canonical tags — this announcement does absolutely nothing to change your daily operations. You are already part of the 75%. This information is mainly a reminder for late adopters.

On the other hand, if you have recently migrated or if HTTP URLs persist in the index, this statistic confirms that Google continues to crawl these old versions. It's essential to check your server logs to ensure Googlebot is not wasting crawl budget on obsolete HTTP URLs that should be permanently removed from the index.

• 75% of Google’s crawl occurs over HTTPS — but this global ratio says nothing about your specific site
• HTTP/2 (available only via HTTPS) enhances crawl efficiency through multiplexing
• HTTP sites are not directly penalized in rankings, but face a technical disadvantage in crawl speed
• Log analysis remains an indispensable tool to check how Googlebot really crawls your site
• HTTPS remains a light trust signal, but its primary impact is technical (crawling, security) rather than algorithmic

Does this statistic really reflect an evolution or a state already achieved?

Let's be honest: this announcement comes late. The massive migration to HTTPS occurred between 2015 and 2018, driven by Chrome warnings, Google's encouragement, and the deployment of Let's Encrypt. As SEO practitioners, we have observed for at least three years that the vast majority of bot traffic is HTTPS.

This 75% ratio is therefore less of a revelation and more of an official confirmation of an existing reality. What's interesting is the residual 25%: it shows that Google continues to actively crawl HTTP sites, either because they have never migrated, or because HTTP URLs persist in the index despite redirects. [To be verified]: Google does not specify whether this 25% includes only fully HTTP sites or also orphaned HTTP URLs from otherwise HTTPS sites.

Does HTTPS really have an impact on ranking or only on crawling?

Google has always been deliberately vague on this point. Officially, HTTPS has been a light ranking signal since 2014 — a “tiebreaker” between two equivalent pages. In practice, the direct algorithmic impact is nearly undetectable. What really matters is the indirect impact.

An HTTPS site benefits from HTTP/2, leading to better crawl efficiency, potentially resulting in a fresher index. It also enjoys positive user perception (no browser warnings), thus indirectly gaining better behavioral signals. But isolating a pure SEO effect from HTTPS? Nearly impossible under real-world conditions. Correlation studies show that HTTPS sites rank better... but this is an evident selection bias: well-managed sites have migrated to HTTPS.

What are the situations where this rule does not apply?

There are a few scenarios where HTTPS creates more problems than it solves. Very old sites with legacy architectures may encounter performance issues after migration (poorly optimized TLS overhead, absence of HTTP/2). Some niche sites in closed environments (intranets, business applications) have no interest in migrating.

More commonly, poorly executed HTTPS migrations — invalid certificate, mixed content, chain redirects, inconsistent canonicals — temporarily degrade SEO. In these cases, it is not HTTPS that is the problem; it’s the hasty migration. But the result is the same: a loss of rankings and traffic. That’s why a HTTPS migration requires a rigorous checklist and close monitoring of logs for several weeks.

What should you check if your site is already HTTPS?

Even if your site is officially migrated, HTTP URLs often persist in Google's index. The first action: download an export of your indexed URLs via Google Search Console (Settings → Crawl → Indexing State → Export) and filter for URLs starting with http://. If you find any, it means Google is still crawling them.

Next, analyze your server logs for at least a week. Look for Googlebot requests to HTTP URLs. If the volume is significant (say, more than 5% of your total crawl), you are wasting crawl budget. Solution: identify these URLs, ensure they redirect properly with a 301 to HTTPS, then request their removal from the index via the URL removal tool in Search Console.

What errors should you avoid during an HTTPS migration?

The classic mistake: forgetting to update internal canonicals. Do all your canonical tags point to the HTTPS versions? The same question applies to your XML sitemap: does it still contain any HTTP URLs? Google follows these signals — if you indicate that the canonical version is HTTP while redirecting to HTTPS, you create an inconsistency that slows indexing.

The second trap: mixed content. If your HTTPS pages load resources (images, scripts, CSS) over HTTP, browsers will display a warning. Google does not directly penalize mixed content, but it degrades the user experience and can impact Core Web Vitals (loading times). Use a crawler like Screaming Frog to detect all HTTP resources on your HTTPS pages.

How to optimize HTTPS crawl for large sites?

On sites with millions of pages, crawl efficiency becomes critical. HTTP/2 helps, but your server must be configured to take advantage of it. Check that HTTP/2 is enabled (test via webpagetest.org or curl) and that your server supports multiplexing.

Another lever: reducing TLS negotiation time. A poorly optimized TLS handshake adds 100-200 ms per request. Enable OCSP stapling, TLS session resumption, and use a CDN with TLS termination at the edge. On a crawl of 100,000 URLs/day, these optimizations can save Googlebot several hours — thus improving the freshness of your index.

These technical optimizations may seem straightforward on paper, but their implementation on a complex infrastructure requires specialized expertise. If your site generates millions of pages or faces crawl budget issues, engaging a specialized SEO agency can save you months and help avoid costly visibility mistakes.

• Export indexed URLs via Search Console and identify residual HTTP URLs
• Analyze server logs to quantify HTTP vs HTTPS crawl by Googlebot
• Ensure all HTTP→HTTPS redirects are in 301 (not 302)
• Check that canonical tags and the XML sitemap point to HTTPS
• Scan the site to detect mixed content (HTTP resources on HTTPS pages)
• Test HTTP/2 activation and optimize the TLS configuration (OCSP stapling, session resumption)