What does Google say about SEO?

Should You Really Stop Relying on Lighthouse and PageSpeed Insights for Core Web Vitals?

Google Dec 10, 2024 ★★★

Barry Pollard, web performance expert at Google, has emphasized the importance of using field data rather than lab data to evaluate a website's Core Web Vitals. This recommendation comes as Google prepares an update to throttling parameters in PageSp...

Domain Age & History AI & SEO Web Performance Search Console

Does Google really treat clustering and canonicalization as two separate processes, or is it all just one mechanism?

Allan Scott Dec 05, 2024 ★★★

Clustering consists of grouping pages that Google considers identical, while canonicalization consists of choosing the best URL among that cluster. These are two distinct and sequential processes.

Domain Age & History Crawl & Indexing Domain Name

What happens when your canonicalization signals contradict each other?

Allan Scott Dec 05, 2024 ★★★

When strong signals like a 301 redirect and a rel canonical point to different URLs, the system ignores these signals and falls back on weaker signals like sitemaps or PageRank.

Domain Age & History Crawl & Indexing AI & SEO Links & Backlinks Domain Name Redirects Search Console

Does Google really juggle 40 different signals to pick the right canonical URL?

Allan Scott Dec 05, 2024 ★★★

Google uses approximately 40 different signals to determine which canonical URL to choose in a cluster of duplicate pages. This number varies over time because certain signals are added or removed.

Domain Age & History Crawl & Indexing AI & SEO Domain Name

Does hreflang really work independently from duplicate content clustering?

Allan Scott Dec 05, 2024 ★★★

Hreflang is a separate system from clustering that allows URLs to be substituted based on user location, even if pages are in the same duplication cluster.

Domain Age & History Domain Name Local Search International SEO

Is Google really about to give trusted sites an hreflang fast-track to indexing?

Allan Scott Dec 05, 2024 ★★

Google is working on a project to increase hreflang adoption by verifying site reliability. If a site implements hreflang correctly, Google will serve appropriate variants more often without systematic verification.

AI & SEO International SEO

Does rel canonical really play a dual role in Google's algorithm?

Allan Scott Dec 05, 2024 ★★★

Rel canonical first serves to put two pages into the same cluster, then if they are clustered together, it also becomes a canonical selection signal to determine which one to display.

Domain Age & History Crawl & Indexing Local Search

Is x-default really functioning as a canonical signal like the others?

Allan Scott Dec 05, 2024 ★★

X-default functions as a canonicalization signal indicating which page to display when localization is unknown. This is different from rel canonical because it does not force clustering, only selection.

Domain Age & History Crawl & Indexing Local Search International SEO

Do 200 Error Pages Really Create Clustering Black Holes?

Allan Scott Dec 05, 2024 ★★★

Error pages served with HTTP 200 status become clustered together by checksum. Pages falling into these clusters escape with difficulty because crawling avoids duplicates, creating a 'black hole' of lost pages.

Domain Age & History Crawl & Indexing HTTPS & Security

Is your redirect chain preventing Google from choosing the HTTPS version as canonical?

Allan Scott Dec 05, 2024 ★★

Complex redirect chains, particularly those alternating between HTTP and HTTPS, can prevent Google from selecting the HTTPS version as canonical if the signals are contradictory.

Crawl & Indexing HTTPS & Security Images & Videos Redirects

Does Google actually prioritize HTTPS in search results, or does it depend on other factors?

Allan Scott Dec 05, 2024 ★★

Google uses several specific criteria to manage the selection between HTTP and HTTPS versions of a page. The principle is to display an HTTPS page only if it is truly secure for the end user.

Domain Age & History Crawl & Indexing HTTPS & Security

Does Google really treat boilerplate translations and full content translations in completely different ways?

Allan Scott Dec 05, 2024 ★★★

Google distinguishes between boilerplate translations (menus, interface) and complete content translations. The former are clustered together, the latter remain in separate clusters because they capture different queries.

Content AI & SEO Pagination & Structure International SEO

Does Google really remove pages faster with a no-index than with a 404 or 410 error code?

Allan Scott Dec 05, 2024 ★★★

An HTTP error code provides a grace period before deindexation in case the error is temporary. A no-index commands immediate removal from the index. Don't use no-index for temporary errors.

Crawl & Indexing HTTPS & Security AI & SEO JavaScript & Technical SEO

Are soft 404 pages really the only ones creating problematic clusters in your index?

Allan Scott Dec 05, 2024 ★★★

Pages served with correct HTTP error codes (404, 403, 503) do not create clustering problems. Only pages returning a 200 code with error content cause indexation black holes.

Domain Age & History Content HTTPS & Security

Can a clear error message really save your crawl budget from clustering disasters?

Allan Scott Dec 05, 2024 ★★

For JavaScript sites unable to send HTTP status codes, displaying a clear and explicit error message like '404 page not found' helps Google detect the error and avoid problematic clustering.

Domain Age & History HTTPS & Security AI & SEO JavaScript & Technical SEO Local Search

Does Google really handle JavaScript redirects to error pages correctly through clustering?

Allan Scott Dec 05, 2024 ★★

Using JavaScript to redirect to a static page returning the correct HTTP error code works because indexing assembles the redirect chain and sees the final HTTP result.

Domain Age & History Crawl & Indexing HTTPS & Security JavaScript & Technical SEO Redirects

Can an empty rel canonical really wipe your entire site from Google's index?

Allan Scott Dec 05, 2024 ★★

An empty rel canonical or one with an unevaluated variable can be interpreted as pointing to the server root, effectively requesting site removal. Google has partial but imperfect validation.

Crawl & Indexing AI & SEO

Can you really control Googlebot News and Googlebot Search separately with distinct meta robots tags?

Martin Splitt Dec 04, 2024 ★★

You can target specific robots by using their name in the name attribute of the meta tag. For example, 'googlebot-news' instead of 'robots' allows you to give instructions only to Googlebot for Google News while allowing classic Googlebot Search to p...

Domain Age & History Crawl & Indexing Discover & News

Can the HTTP X-Robots Header Really Replace Your Meta Robots Tag?

Martin Splitt Dec 04, 2024 ★★★

Instead of using an HTML meta tag, you can use an HTTP header called 'X-Robots-Tag' which can contain exactly the same values as the meta robots tag, offering an alternative way to control indexation.

Crawl & Indexing HTTPS & Security Images & Videos

Should you manage a separate robots.txt file for each subdomain?

Martin Splitt Dec 04, 2024 ★★

Each subdomain can have its own robots.txt file. For example, shop.example.com/robots.txt is valid and functions independently from the main domain's robots.txt.

Crawl & Indexing AI & SEO JavaScript & Technical SEO Domain Name PDF & Files