Google used to be about transporting you around the open web and connecting you with all the weird, wonderful stuff the internet has to offer.
Not anymore. If it was up to Google, you’d never need to leave its growing internet real estate. It’s a scary proposition for just about everybody but Google.
Between fast-loading AMP articles from major news brands hosted in its domain, full pages of information scraped from outside sites that don’t require you to visit them, basic shopping functions built into ads, YouTube, and a host of other features, the Google-verse is more of a digital walled garden than ever.
The most recent addition comes in the form of a report that the company is altogether.
Google has always had these ambitions to one extent or another. Former Google CEO Eric Schmidt said way back in 2005 that the platonic ideal of a Google search should ultimately yield just one result, and that result would be the simplest answer to the query, no clicks needed.
The company has made various moves towards this idea in the years since, but it only really became aggressive about it when arch-rival Facebook’s own drive towards its own enclosed platform lit a fire.
The benefits of operating within a regulated walled garden are obvious. The open web tends a wild and messy place, whereas closed platforms allow companies to control, track, and potentially monetize every part of their user experience and behavior.
Google’s a bit different because it primarily operates in the web at large. Its core business is built around organizing and indexing the internet’s chaotic expanses. Its networks place ads on thousands of different sites, reaping billions of dollars in the process.
The company justifies its new crusade as a boon to consumers, who obviously aren’t big fans of slow-loading pages, extra navigation, and annoying ads. And six in ten Google users say they want more results they don’t have to click.
Google puts *their* code layer (AMP) in bold, dark type with icon while commoditizing news brands in gray type. Not as bad as Facebook but.. pic.twitter.com/JpOUleYCbk
— Jason Kint (@jason_kint) July 22, 2017
But that convenience doesn’t come without a cost. Brian Warner, founder and CEO of CelebrityWorthNet.com, understands perhaps more than anybody the power of Google’s wall-building.
Warner started to notice the content from his site appearing directly on search results pages in 2012. Two years later, he got an email from Google asking to scrape all of his data, which he turned down. Another two years after that, Google did it anyway, and the impact was catastrophic.
“It was extremely painful, it was extremely devastating,” Warner said. “We got to a point where our traffic was down 85 percent from a year or two earlier.”
Search for the net worth of any celebrity at random today—let’s say, James Earl Jones—and you’ll get the number ($45 million) and a short biographical blurb pulled from CelebrityNetWorth.com with credit and a link. Below that is a panel of questions commonly asked about the Star Wars actor (“Is James Earl Jones alive or dead?” reads one) and dropdown answers pulled from sources ranging big news sites to personal blogs (“James Earl Jones is alive!” E!’s excerpted headline rejoices.)
More questions load every time you click, and it’s easy to stray from related topic to topic without ever leaving Google.
The growth in the use of these sorts of built-in tools in the past few years has been dramatic. A recent report from marketing agency Stone Temple found that half of all Google search results now come with some form of information hosted within the site—and three in ten with so-called “snippets” in particular. As of January, these excerpts appeared more than 50 percent more often than they did just a year and a half earlier.
Google’s mobile experience has become even more insulated from the rest of the web. Some secondary results pages are now entirely self-contained; they list relevant information directly on the page and link only to AMP articles, YouTube videos, and other results pages in the same format. It’s not hard to imagine how this set-up might be fleshed out into an entirely closed information network one day.
There’s also a steady stream of more subtle indications of Google’s inward pull appearing every day—features like on-site hotel booking, restaurant menus, spa appointment tools, and dropdown recipes to name just a few.
These tweaks might sound minor, but Google’s position as the web’s central nervous system means they can have big impact on smaller businesses that orbit it.
In the long run, there seems to be a pretty glaring hole in this plan. That is, as Google likes to reassure wary publishers, it’s not in the content business.
The company ultimately relies on reference sites like Wikipedia, IMDB, Fandango, and the CIA World Fact Book to compile and update the information it uses.
If Google continues to choke these sites out, what incentive will there be for new ones to come along?
“CelebrityNetWorth is a website I created because I thought it needed to exist and I liked it,” Warner said. “There’s no way I would go about that today.”
The information Google plasters on results pages isn’t immune to mistakes, whether because of the quality of the site from which it was pulled or a data misalignment on Google’s part.
How often do you come across a “featured answer/snippet” in Google that is just wrong? I’m seeing it more, wondering about refreshes.
— Julie F Bacchini (@NeptuneMoon) May 15, 2017
This seems like Google are accidentally overreaching and providing bad/outdated information in their Featured Snippet or whatever this is pic.twitter.com/xgIBssTZL3
— Simon Wheatley (@simonwheatley) May 19, 2017
Funny AI blunder: this google query returns a text snippet but cuts off the very next line “Yes, I just made all of this up” pic.twitter.com/ASUSzvGP1O
— Ron Shukartsi (@Rig8_Ron) July 22, 2017
These sorts of functions fall under the province of Google’s much-touted “Knowledge Graph,” a program it built in 2012 to populate its results pages with aggregated content. The system has shifted search’s focus from just words to more complex concepts that text represents. The engine powering it is constantly vacuuming up new data from all over the web.
It’s an impressive tool, but the quality of the information it stores is ultimately only as good as that of the web around it.
And as Google puts the squeeze on publishers, it risks deteriorating it.
For now, though, that concern doesn’t seem to be at the top of Google’s priority list as it works on swallowing more and more of the internet.