What Has Pandu Nayak Said About Page-Level Rankings? A Tales from the Query Post

This article is part of a new series for my blog called “Tales from the Query,” where I look at a random query in GSC and write a blog post about it.

Today’s query is: “pandu nayak page level rankings.“

I already have an article that ranks for this query, “What Today’s Google Search Ranking Volatility Says About Tomorrow’s SEO Strategies”:

But it’s a little dated (2024), and I think this query’s intent deserves its own page.

So let’s get to it.

Nayak describes scoring individual pages, but the evidence also shows that broader signals can influence those rankings.

But first off, who is Pandu Nayak?

Pandu Nayak is a computer scientist and Google Fellow and Chief Scientist for Search, previously serving as a vice president leading Google’s Search Quality team, according to his website’s biography.

His work focuses on language understanding and improving how Google retrieves and ranks information.

Now, let’s look at what Pandu Nayak has said about page-level rankings based on some quotes we can pull, with my highlights.

Antitrust testimony: each matching page gets a score (October 18, 2023)

Perhaps the clearest exchange about page-level rankings comes from the DOJ antitrust trial on page 6,401 (PDF page 10):

Q: And the core algorithms are critical, because if they don’t bring up the right documents, then even RankEmbed doesn’t go back and redo the work that the core system is supposed to do, right?

A. I mean, it could, but — so right now it doesn’t, yes.

Q: “And at the end, each page that matches a query, it gets a score?”

A: “Yes.”

Q. And then Google sorts the scores, and that’s what’s used in part for Google to present to the users?

A. Yes.

He thus confirms that Google sorts the page-level scores and uses them, in part, to determine what users see.

Note that the discussion concerns scoring web results for a query, not whether every input originates at the page level.

Google blog: understanding the concepts in a page (February 3, 2022)

In “How AI powers great search results,” Nayak writes about neural matching:

“Neural networks underpin many modern AI systems today. But it wasn’t until 2018, when we introduced neural matching to Search, that we could use them to better understand how queries relate to pages. Neural matching helps us understand fuzzier representations of concepts in queries and pages, and match them to one another. It looks at an entire query or page rather than just keywords, developing a better understanding of the underlying concepts represented in them. Take the search “insights how to manage a green,” for example. If a friend asked you this, you’d probably be stumped. But with neural matching, we’re able to make sense of it. By looking at the broader representations of concepts in the query — management, leadership, personality and more — neural matching can decipher that this searcher is looking for management tips based on a popular, color-based personality guide.”

He explains that the system identifies the concepts represented by queries and pages, helping Google retrieve relevant documents. This supports page-level semantic understanding, but it is not a claim that ranking uses only page-level signals.

Google blog: site-level demotions (June 10, 2021)

In his post “Improving Search to better protect people from harassment,” Nayak writes:

“One such example is websites that employ exploitative removals practices. These are sites that require payment to remove content, and since 2018 we’ve had a policy that enables people to request removal of pages with information about them from our results. 

Beyond removing these pages from appearing in Google Search, we also used these removals as a demotion signal in Search, so that sites that have these exploitative practices rank lower in results. This solution leads the industry, and is effective in helping people who are victims of harassment from these sites. ”

The context is websites charging people to remove harmful content. Google could remove particular pages and use those removals to demote the sites involved.

This is useful context because Nayak publicly describes a ranking intervention extending beyond an individual removed page.

Antitrust exhibit: document-quality scoring (January 31, 2025)

Notes from a call with Nayak — DOJ exhibit PXR0357, page 1 — describe Q* as:

“Q* (pronounced “Q star”), Google’s measure of quality of a document;”

The notes also say PageRank contributes to page-quality signals:

“PageRank, one of Google’s original signals, continues to be one of the factors that goes into the quality signals for pages.“

They also describe an internal debugging interface showing scores for individual results:

“[Redacted] containing a table with list of 10 blue links and corresponding score for each top-level signal as well as corresponding total score “Final IR” across all signals…”

Remember, this is a litigation exhibit containing call notes, not a verbatim testimony transcript.

Creator Summit: the disputed page-level statement (October 2024)

In the closed Creator Summit, attendee Morgan McBride recalled Nayak saying:

“There are not site wide penalties, only page level queries.”

That wording comes from McBride’s recollection in an interview transcript. I haven’t located an official recording or verbatim transcript of Nayak’s summit remarks to independently verify it.

Also note that the interview is critical of Google and reflects a publisher’s experience of substantial traffic losses.

A separate firsthand account, by Going Awesome Places’ Will Tang, summarizes the message more cautiously:

“They stressed that page-level signals are the majority of what is used when serving search queries.”

Tang’s account describes a panel involving Nayak, Elizabeth Tucker, and Danny Sullivan. This is Tang’s summary of the discussion, not an individually attributed Nayak quotation.

Google’s explicit clarification: page-level ranking includes site-wide signals

This is from “Google’s guide to Search ranking systems,” not a Nayak quote, but it directly resolves the central distinction:

“Our ranking systems are designed to work on the page level, using a variety of signals and systems to understand how to rank individual pages. Site-wide signals and classifiers are also used and contribute to our understanding of pages. Having some good site-wide signals does not mean that all content from a site will always rank highly, just as having some poor site-wide signals does not mean all the content from a site will rank poorly.”

So, while Google ranks individual pages, it also has site-wide signals. This is an important takeaway: a page’s ranking can be influenced by site-wide signals, so evaluating its content alone may not explain its performance.

More takeaways:

In short, the evidence we’ve reviewed from Nayak and Google supports us saying that Google scores individual pages for queries while also using information and signals beyond those pages.

Here are some additional takeaways from the sources:

• Individual pages are the units Google scores and ranks. Nayak confirmed in his antitrust testimony that each page matching a query receives a score. While that supports describing ranking as page-level, it doesn’t establish that every signal comes from that page.
• Relevance involves meaning, not just matching keywords. Nayak’s explanation of neural matching describes Google interpreting the concepts represented by a query and a page.
• Page quality matters alongside topical relevance. Nayak’s testimony identifies page quality and reliability among Google’s considerations. Addressing the right topic alone does not fully explain why a page ranks.
• Page-level ranking can incorporate broader signals. Nayak’s blog about harassment describes using removals to demote sites with exploitative practices. Therefore, his published words don’t support treating each page as entirely independent of its website.
• Ranking combines multiple systems. Nayak discusses document content, topicality, quality, localization, and Navboost. Explaining rankings through a single factor risks overlooking how those signals interact.
• The creator-summit remarks require careful attribution. The strongest claims about there being no site-wide penalties come from attendees’ recollections, but these are reported remarks rather than independently verified quotes.
• The central distinction is between what gets ranked and what informs the ranking. Google ranks pages, sure, but the information used to evaluate those pages can extend beyond their individual content.

Well, that explains the query “pandu nayak page level rankings.” I hope you’ve found it helpful, whoever is out there searching it.

Lastly, if you need a hand with your SEO/GEO strategies, feel free to contact me. I’m an independent consultant helping brands and agencies.

Until next time, enjoy the vibes:

Thanks for reading. Happy optimizing!

Related posts