Google promises to adjust search algorithm to favor 'people-first content'

Well now there's an idea


In what may be a major shift, Google plans to revise its search algorithm to rank content crafted primarily for people rather than content designed to impress web crawlers and indexing bots.

"Next week, we'll launch the 'helpful content update' to tackle content that seems to have been primarily created for ranking well in search engines rather than to help or inform people," said Danny Sullivan, public liaison for Google Search, in a blog post on Thursday.

"This ranking update will help make sure that unoriginal, low quality content doesn’t rank highly in Search, and our testing has found it will especially improve results related to online education, as well as arts and entertainment, shopping and tech-related content."

The Chocolate Factory regularly revises its search algorithm, for better or worse, in response to changing business priorities. This also helps defend against search engine optimization that bends or breaks company rules to maximize ad revenue rather than promoting quality content.

But even so, the search giant's results of late appear to have attracted more criticism than usual.

Last December, for example, dissatisfaction with low-quality results from queries related to programming code prompted a discussion about the rise in low-quality code snippet sites. These "code snippet aggregators" show up on Google search results pages (SERPS) when people search for particular bits of programming code or error messages and many appear to be scraped from sites with actual programming communities like Stack Overflow.

And the following month, YCombinator partner Michael Seibel tweeted, "A recent small medical issue has highlighted how much someone needs to disrupt Google Search. Google is no longer producing high quality search results in a significant number of important categories."

Not long after, Rob Beschizza, from BoingBoing.net, advised searchers to add the word "reddit" to any given search as a way to guard against inauthentic content – evidently, Reddit has managed to resist being overrun with spam posts. Beschizza more recently referred to Google Search as "an affiliate revenue listicle trashcan."

And rivals like Brave observe, "results pages in Big Tech search engines like Google are often cluttered with ads and automated content (or 'SEO spam') from marketers trying to game the system and increase the rank of their sites."

To some extent it has always been thus. But this update, arriving in the wake of vocal criticism, not to mention Brave's launch of a rival search service, suggests Google is making a larger-than-usual course correction to defend its 91 percent global market share.

Sullivan earlier this year contributed to that effort by pointing out a study that found Google's rival Bing is worse when it comes to misinformation.

Now Google's focus on ensuring that people see "more original, helpful content written by people, for people, in search results" will be carried out by machine learning.

"This update introduces a new site-wide signal that we consider among many other signals for ranking web pages," Google explains on its Search blog. "… This classifier process is entirely automated, using a machine-learning model. It is not a manual action nor a spam action."

Google's new cruelty won't necessarily banish all cynical clickbait traps larded with chumboxes. The ad biz acknowledges that some sites with unhelpful content could still rank well on SERPs if those sites present sufficiently "people-first" material in an algo-pleasing way. But the tech giant says it will "continue refining how the classifier detects unhelpful content and launch further efforts to better reward people-first content." ®


Other stories you might like

Biting the hand that feeds IT © 1998–2022