Press "Enter" to skip to content

Necessary Evil: Pushing Navboost, Lets Push Google’s Top Ranking Signal – Part Three

By Brett Tabke on April 15, 2025

Views: 135

Pushing NavBoost – In Three Parts

  1. NavBoost 2025: What’s it all about.
  2. Signals Everywhere: what user data can we track on the Serps?
  3. Pushing NavBoost: Can we Influence it? – Of course we can – it’s our job!

Yes, but can We Influence NavBoost to Rank Higher?

So far, we’ve mapped out what NavBoost is and the signals it has available to watch. But now we get to the real question that every SEO eventually asks – can we actually push this signal at all? In this final installment, we’ll look at what’s possible, what’s already being done, and how SEOs are using intentional behavior patterns to influence rankings. It’s not theory anymore. It’s field-tested experimentation that blurs the line between user simulation and signal shaping.

Lets backup and hear from Marcus again over at Links-R-Us: “Yes, we have clear one-to-one data over the last nine months showing rankings being influenced by clicks. People told us that we needed to think in terms of large scale in order to make this work, but that is just not the case in the long tail where conversion rates are highest. We saw results with as few as a dozen clicks on Serps a day over six months. We can’t wait to see six more months of data.”

(edit 4-15-25: from Marcus: “…where we have seen the largest increase is from page two to page four in getting rankings up to page one.“)

How did they do it?

  1. They wanted to look as close to a human browser as possible. (more on that below)
  2. They clicked a few times on their url on known kw Serps every day from multiple IP’s
  3. They went back during the day and did other searches, in order to look more human
  4. 50% of the IP’s were logged into Google accounts.

All of this was automated with browser drivers. They went so far as to mouse around and hover over some results before randomly clicking on their links. They also used browser VPN services to rotate IP’s every few clicks.

While, theirs was quite an effort, new browser tools are opening this up for anyone. Things like website test programs work well for clicking links and behaving like a user. You can also try “Browser Use” (website) AI agent .

One aspect that Links-r-Us played around with was with major click hit’s on long-tail keywords for breaking news stories. They were trying to find out of Instant Glue could be pushed. Another screen shot from the USDOJ vs Google antitrust trial:

They were unable to see any movement in the urls they attempted to lift in the short term, but did see lift in the nine month results.

We don’t know what we Don’t Know

Top things we really don’t know:

  • IS Scoring. Repeatedly in the trial documents, IS (Information Score) is referenced. It appears to be in reference to quality raters aggregate score. We really don’t know how that all fits together. There are only vague references.

What we don’t know is a huge chunk of how it all fits together, the weights given to each metric, or how the various filters like Instant Glue are applied.

Do the above revelations, and the previous installments of this series, leave us any better off than we were before?

While researching this issue over the last couple months, I dug through maybe 200 websites/pages that had references to NavBoost and related terms. What I found was some killer information (see: Resources below), and then some flat-out wrong information. No article of this size would be complete without pointing out some repeated misinformation about NavBoost that is out there:

  • Widely circulated is this one related to: “On-Page Engagement” of a destination page. Some how people got the idea that NavBoost had something to do with the quality or EEAT of the destination url of a Serp click. I can not find anything like that in the documents we have about NavBoost. That is not to say, there might not be a component to that through the usage of Google Analytics. Like NavBoost itself, it has long be suspected that Google leverages Google Analytics data from websites to find high performing pages or other site data. It sure seems like there is a direct and disproportionate number of number one to number three ranking pages, that have Google Analytics installed on the site. It is disproportionate because lower ranking pages more often do not have GA installed on their site. (hmmm, raise eyebrows – yes I know Google says it does not use GA data, but everything we know about Google is that they use every scrap of data available to them. For them not to use GA data? It doesn’t pass the sniff test. I simply can’t imagine a scenario where Google has a big ‘ol hunk of data like this and not turn an LLM model loose on it for training.)
  • “Bounce Rate is everything“. If I want to know the weather for the day and I search Google for the weather, and I click on weather.com and then go back to Google for another search, does that mean that Weather.com is a failure because of a high bounce rate? No, but it does mean that the time in between should fall within some norm for that site. Naturally, some pages/clicks will have higher dwell time that others. I believe bounce-rate is highly over rated. In the world of multi-tab browsing, it is a very noisy signal.
  • “Sites with Good UX“: This has nothing to do with NavBoost perse. Like on the page engagement, it indirectly affects it, but there is no direct metric for UX experience within NavBoost that we are aware.
  • “Optimize your site for Navboost“. You can’t. Other than a back-button-breaker, there is little you can do on your website that will directly in any way affect NavBoost. Remember, it is first and foremost a click metric.

In Conclusion

Personally, having watched Google for 27 years, I would put the make up of the Algo as:

But in one sense, that over sells the concept because Google has all these filters and hoops that a page must jump through to get to the ‘final rankings’. I think of it like a bracket for March Madness with each step removing more and more results to get to the final rankings. Google talks in terms moving a page from thousands of results, into a ‘green circle’ where the final rankings are assembled. NavBoost probably isn’t applied until we get to that final stage. Each step along the way as they refine the ‘data set’ of appropriate pages that match the query in various degrees gets a different “algo” or “filter” applied. In that sense, NavBoost isn’t even around for the first stage of results refinement, but it sure ends up being the Judge and Jury later on.

So by the conclusion of this series, it is evident that NavBoost is not merely a bolt-on ancillary feature of Google Search. Rather, it serves as the evaluative layer that activates after all other elements have been sliced, diced, filtered, and organized. It disregards the quality of your content, the number of backlinks, or the speed of your page, unless these factors impact user behavior on the search engine results page (SERP). While we may never fully comprehend every aspect of the system without more data, we have gained this understanding: NavBoost mirrors actual user actions rather than what we presume they ought to do. For SEO professionals, the way forward does not involve speculating about the contents of the Google black box. Instead, it requires an understanding of how genuine behavior influences SERP visibility, utilizing that knowledge to inform strategy, conducting intentional tests, and operating more efficiently within a system that monitors every action. The final bottom line is simple : yes, you can push NavBoost. Thanks!

…pssst: Send me your best NavBoost stories. brett at this one .com

Resources

Views: 0

NavBoost Links and Resources

Published in Google