Blog: SEO
Businesswoman on rooftop using smartphone to do a Google search

Does Google Rank Webpages or Websites?

Avatar for John Locke

John Locke is a SEO consultant from Sacramento, CA. He helps manufacturing businesses rank higher through his web agency, Lockedown SEO.

Does Google rank webpages or websites?

It’s an interesting question, and a January 2021 article by Search Engine Journal suggested that webpages are what Google evaluates, not the websites as a whole. The author even showed a now deleted tweet from Google representative John Mueller as proof.

However, noted SEO consultant Victor Pan, who is currently Head of Growth at HubSpot, said there is more nuance to how Google predicts whether a page will be high quality or not.

In a Twitter thread, which begins here, Pan talked about the nuance involved in webpages vs. websites when it comes to ranking.

On a side note, we believe that getting the right content on a page will improve the chances of the page ranking. I found it interesting that the history of a website can have a lot of influence on the potential ranking of future pages.

Nuance in Google Predicting Page Quality of New Content

In his thread, Victor Pan noted that websites with a history of selling links or with old manual penalties would be less trustworthy to the Google ranking algorithm, no matter how good their content might be. Conversely, websites with a long history of satisfying users, and producing good high-quality content are more likely to have future content ranking well. Google seems to favor sites that put in the effort over a long period of time, with proven signals that users are happy with the site content.

Core Web Vitals are also a predictor of whether a site will be high-quality, in the absence of user signals related to a website’s content, according to Pan.

Site content, page experience, and the history of user signals work together to create a cohesive story about a website. Creating content around a specific topic tells Google that a website belongs in a specific category. Google makes a note of that topical knowledge, and measures the history of whether a website has a history of content that satisfies users. The website as a whole has an influence on individual pages, and their ability to rank.

To sum up, Google uses predictive models to tell whether a particular new web page should rank high based on factors like the history of the site, and the overall perception of the site, not just domain authority or the backlink profile.

What Did Victor Pan Say About Websites vs Webpages?

Hi folx. Covid times sure fly. Google ranks webpages, not websites. 
Has a nice ring to it eh? Let’s find talk about nuance. It’s not that simple. Usually I’d grab links and citations, but hey it’s that kind of day and this dad needs a break. - @victorpan

Victor’s thread is referencing a John Mueller statement and Search Engine Journal article centered around “Google ranks webpages, not websites” and his thread shows there is more nuance to ranking.

This is a really well written piece. https://searchenginejournal.com/google-ranks-webpages-not-websites/393425/amp/ honestly I am a fan.  As an industry we need to move away from using a PageRank proxy (DA) to evaluate websites and backlinks. Ask: do our audiences overlap? Not: is their DA high? But it oversimplifies nuance. - @victorpan

Domain authority, as measured by SEO tools, is a simplistic way to measure the utility of a website.

You could argue that pornhub insights e.g. https://pornhub.com/insights/corona-virus should not be filtered by SafeSearch. Yet when you add ?safe=on as a query parameter it disappears from search. It’s not uncommon for whole websites to be filtered out. Defining websites is tricky. - @victorpan

The link in question was an analysis on how Coronavirus affected traffic of the site mentioned. The page has no controversial content, but the overall website is known for adult content. Victor is making the point, the reputation of the website affects the ranking of a page, even though Google representative John Mueller said that “Google ranks webpages, not websites.” I needed to give you the overall context of this tweet, and how it fits into the concept of how Google ranks search results.

Google ranks webpages, not websites - John Mueller

The now-deleted John Mueller tweet from the Search Engine Journal article.

Bought a domain that was previously penalized for peddling links? I’m sorry there is no remedy but to wait until Google’s systems process through the site-wide penalty. This is not a page-level-we-can-fix-it issue. This is a whole site ranking issue. - @victorpan

Victor says here Google cannot always rank webpages without evaluating the overall history and context of the website. If you buy a penalized domain, it is essentially useless, no matter how good the new content published on that domain. Even normal, non-penalized websites have a data history that Google takes into account.

Those are all penalty-related, are there any non-penalty related factors and nuance? Yes. A website’s history gets used as a ranking signal fill-in for net new content from a website where signals are missing. - @victorpan

When real signals are missing, Google uses the history of that website to fill in as proxy data. This creates a lot of nuance about the context of a website.

If you’ve ever experienced, dang anything I publish ranks in the first page of Google, in a few hours then you know what I am talking about. The only backlinks that article has at that moment are from scraper sites and content farms, plus or minus a few internal links. - @victorpan

What Victor is speaking about here is Google trusting a site to publish good content. When a site has a track record of satisfying users with their content (however you believe Google measures that), content that they publish tends to rank higher than their competitors automatically. This is exactly the “big brand” effect that Google has aimed towards since at least 2008. They want to rank websites that they already trust, that have a proven track record of quality.

Let’s talk about the future, too. Web Core Vitals will rely on Chrome User-Experience Report, refreshed on a rolling 28-day average. What do you do with new pages from a site you do not have data on? Use an aggregate as a predictor. https://google.com/amp/s/www.seroundtable.com/amp/google-slow-fast-granular-speed-urls-30283.html -@victorpan

Page experience is a predictor of overall site quality, more on that in a minute.

So the argument that pages make the website... well that’s common sense. 
But the argument that a page on a new website has the same chances of ranking as another page from the HubSpot is just not true. Site averages get used for net new content in the absence of data. - @victorpan

Despite what you’ve heard, the playing field is not meant to be even. Big brand websites with a long history of quality content, good page experience, and all the signals that this is a large organization, Google is going to trust that 100 times out of 100 before it ranks a brand new website above it. The risk is small that the big-brand, historically-quality website will deliver a search result that does not satisfy users.

I absolutely agree that the focus on SEO needs to be on the page or set of pages. The history of your site will take your organic traffic to new heights with that approach. It’s a virtuous cycle. Yet the reality is not all sites have good history. - @victorpan

Building a good history for your website is a long project. Google wants to reward the websites that do the difficult things, not the easy things. Focus on long-term initiatives, not only short-term quick wins.

Google categorizes websites and their pages when there is enough data. The categorizations can be a positive ranking signal, a filter, or even a penalty. Google ranks pages - using site data. When Google de-ranks websites, the pages don’t matter. - @victorpan

This is an interesting insight. Google absolutely categorizes websites when it has enough information. What topic does your website revolve around? Does your site give satisfying content and a positive page experience for users? Worth noting that the “filter” doesn’t necessarily mean your content is “bad” or low-quality. If your content goes against the grain of the rest of the site’s content, Google’s algorithm may not rank it, because it doesn’t historically fit that category. Finally, penalties tend to only apply to sites who have tried to manipulate search rankings using deceptive tactics.

Page Experience and SEO

Page experience is another ranking signal, according to a November 2020 post on the Google Search Central blog. Content is important, and your site itself must have a long track record of quality and making searchers happy. But this excerpt from Google’s post on page experience is something you should also meditate deeply upon.

Page experience is a set of signals that measure how users perceive the experience of interacting with a web page beyond its pure information value, both on mobile and desktop devices.

And also on that same page:

While page experience is important, Google still seeks to rank pages with the best information overall, even if the page experience is subpar. Great page experience doesn’t override having great page content. However, in cases where there are many pages that may be similar in relevance, page experience can be much more important for visibility in Search.

Page content, the long-term track record of your website, and site experience are all things to consider when talking about your SEO strategy.

Does Your Company Website Need SEO Help?

If your industrial or manufacturing website needs help with SEO, the team at Lockedown SEO is here to help. Reach out to us about your project, and we can make some recommendations that will help you rank higher for your target keywords, reaching your ideal clients.

Avatar for John Locke

John Locke is a SEO consultant from Sacramento, CA. He helps manufacturing businesses rank higher through his web agency, Lockedown SEO.

Join the Conversation

Your email address will be kept private. Required fields marked *.