Insights
What Google revealed about queries and ranking
The Value of Queries
"Let's start with some background... A billion times a day, people ask us to find documents relevant to a query. What's crazy is that we don't actually understand documents. Beyond some basic stuff, we hardly look at documents. We look at people. If a document gets a positive reaction, we figure it is good. If the reaction is negative, it is probably bad."
—Eric Lehman, former software engineer at Google
The data on query clicks is vital for Google to determine which relevant pages to present for any query. Judge Mehta’s ruling in the antitrust trial stated:
"The sheer magnitude of Google's query volume, or scale, compared to rivals is startling."
Google uses this data to:
- Crawl additional websites
- Expand the index
- Re-rank the SERP
- Improve the “freshness” of results
Narrowing Down The Search
We understand how important Google’s index is in retrieving pages that match a query.
With millions of potential results, the general search engine (GSE) must narrow the volume of links to just a few thousand.
Those results are ranked at a more granular level, with scoring determining which results are shown to the user.
Interestingly, Eric Lehman testified in court:
“The middle problem of figuring out what are the most relevant pages for a given query in a given context still benefits enormously from query click information.
And it's absolutely not the case that AI models eliminate that need or supplant that need.”
Judge Mehta agreed:
“Google, however, continues to rely on large volumes of user data at every step of the search journey, and no witness, even from Google, testified that LLMs had sufficiently advanced to supplant user data.”
The Role of Click and Query Data
While click data is not particularly important for indexing, query data is essential.
GSEs need to ensure their index covers frequently entered queries.
A notable signal used is Navboost, which memorizes user click data to pair queries with documents, allowing Google to recognize patterns in which documents are clicked for multiple queries.
Types of Queries
Google categorizes queries into three types:
- Non-commercial Queries: These are informational and make up 80% of Google’s queries. They are not monetized with ads.
- Commercial Queries: Queries that express a user’s commercial intent, making up the remaining 20%. These are often accompanied by ads.
- Navigational Queries: These can be either commercial or non-commercial and involve users searching for a specific website, like “Amazon.”
Interestingly, Google’s top five queries are all navigational, and nearly 12% of all queries fall into this category.
Ranking in Google Search
Google uses query data extensively when ranking search results.
Documents released in the antitrust trial, such as "The 3 Pillars of Ranking," shed light on how Google values user interaction signals like click-through rates and time spent on a website.
“As I mentioned, not one system, but a great many within ranking are built on logs. This isn’t just traditional systems, but also cutting-edge machine learning systems—RankBrain, RankEmbed, and DeepRank.” [Source].
I discussed ways to improve search rankings in my last newsletter, so please check that out for the background.
Notes on Retrieval and Ranking
Humans—and their queries—are imperfect.
Google uses user data to improve search results, such as identifying misspellings or using synonyms.
The system scores potentially relevant results using various signals (including user data) to determine the ranking order.
Improving Rankings: Key Strategies
1. Understand User Intent and Context
Prioritize research to understand the intent behind user queries.
Create content that directly addresses user needs and questions to improve relevance.
Intent needs to move beyond transactional, informational, and navigational to active, passive, and latent intent analysis.
If search intent changes, adjust your content to align it with what site visitors seek.
2. Monitor User Behavior and Feedback
Use analytics tools to analyze user behavior on your pages.
Regularly review search query data in Google Search Console (GSC) to identify the keywords driving traffic and adjust your content to align with user intent.
Consider adjusting page elements like calls to action or mobile optimization to improve click-through rates and time spent on site.
3. Regularly Update Content
SEO is an ongoing process. Google favors fresher, updated content, and regularly updating your site can boost your rankings.
Diversifying your content types (articles, videos, infographics) can increase visibility and attract a broader audience.
4. Encourage Feedback and Engagement
Engage with your audience through quizzes, polls, surveys, or feedback forms.
By doing so, you foster a deeper relationship with your audience and learn directly from them, which can positively impact key ranking factors.
By taking these steps, you ensure that your content stays relevant and competitive in the ever-changing search landscape.
2. Highlights from the August Core Update
1. It's done
The August 2024 Core Update finished rolling out on Tuesday afternoon—earlier than expected.
According to a poll run by Barry Schwartz, few sites saw full recoveries from the HCU.
His poll results showed:
44% had rankings/traffic drops
27% had rankings/traffic increases
29% had no change
2. Update notes, SEO strategy, and unique challenges
This video update from Glenn Gabe contains several thought-provoking notes on the August core update.
Glenn also addresses:
* Adjusting your SEO strategy based on observed changes and competitor behavior
* YMYL sites and their unique challenges
* Exercising caution with AI-generated content
* Diversifying content formats to maintain engagement
Glenn's observations are worth 15 minutes of your time.
I hope these insights help. For more tips, follow me here.