A recent study says that Google’s universal search harms users—and that claim has implications for journalists.
While the study was conducted by two well-respected academics, it was also funded by Yelp Inc., one of Google’s competitors. The authors are Michael Luca, from Harvard Business School, and Tim Wu, a Columbia Law School professor most famous for popularizing the term “net neutrality.”
The study doesn’t criticize Google’s “organic search”—Google’s standard search method, based on an algorithm that aims to match users with the information that best meets their needs.
Instead, it focuses on Google’s universal search, a feature that’s been introduced and gradually modified in recent years. The feature prioritizes Google Plus selections and leaves out large numbers of search results completely—and the overall results are less helpful, Wu told a Business Insider journalist:
"Some of the Google Plus stuff looks a little spammy. Why does the most famous restaurant in New York have three reviews, you know?” Wu said. “My more-general thesis is, wherever Google Plus goes, bad things follow. There’s a war between what I would say is the true soul of Google and Google Plus.
“So I’m kind of calling on Google to be more Google-y, and I’m fighting for Google over Google Plus,” Wu added.
Wu and Luca conclude that universal search does a disservice to both users and competitors in the process:
The easy and widely disseminated argument that Google’s universal search always serves users and merchants is demonstrably false. Instead, in the largest category of search … , Google appears to be strategically deploying universal search in a way that degrades the product so as to be slow and exclude challengers to its dominant search paradigm.
Here’s how it works: When you do a Google search, you’ll often see a boxed set of results (try [your city] and “restaurants,” for example). This “3 Pack” or “7 Pack”—so-called because it features three or seven businesses, with an option to click for more results—prioritizes results from Google Plus reviewers, downgrading or excluding other review sites with more numerous reviews—reviews which, in many cases, provide better results.
In one of the study’s examples, a search for a pediatrician in New York City using Google’s universal search returned only 31 total reviews. When the study’s authors modified the search to use Google’s own organic algorithm, they received 719 results, with other review sites appearing higher in results.
While the implications might not be so bad for an individual search—you have dinner at a good restaurant rather than a great one, or you take your car to a lesser auto mechanic, for instance—the study says that there’s a cumulative effect:
Consider, for example, a consumer who is misdirected and ends up at a bad restaurant; or the parents who are looking for a topnotch pediatrician but because of search degradation, patronize a subpar practitioner. The harm caused by such misdirection when it occurs, will vary, but is undeniable in the aggregate.”
The resulting injustice to both users and competitors is exactly the kind of problem that led the Federal Trade Commission and the European Union to investigate Google’s trade practices. Although the FTC has dropped its investigation, Google has until August 17th to respond to the EU’s charges. Journalists are in danger of magnifying those injustices if they rely too heavily on Google, or any search engine, for that matter.
Let’s say you’re looking for an expert for a news story—an attorney, or a doctor. If you begin with a Google search and depend on those results, you’re probably not going to find the most respected local expert in that field. What’s more, you’ll be increasing the false impression created by Google’s initial results by disseminating them further, and by, in effect, endorsing the idea that your source is a leading expert.
To avoid this perpetuation of bias—and to help your story’s accuracy by providing you with sources who are more likely to be the most knowledgeable in their fields—you can check more specialized search sites. For instance, in the case of the medical expert or doctor, you can start with sites such as ZocDoc or AMADoctorFinder or CastleConnolly.com, a site that provides information based on peer reviews and publications.
Google is incredibly helpful; like other search engines, it provides a useful filter. But its biases are only one of its problems, which is why experienced journalists learn to use specialized filters, offering more relevant and accurate information. For instance, you can turn to websites created specifically for journalists, such as the Journalist’s Toolbox, maintained by the Society of Professional Journalists. That site provides links to databases, experts, and hundreds of resources. You can also check The Responsible Journalist's Resources page, which provides other lists of resources that journalists have found useful.
We’ll be watching the European Union’s case against Google with interest, but in the meantime, Wu and Luca’s study provides a timely reminder to journalists to broaden their sources of information. If you’re too dependent on any one person or place—or search engine—you’re more vulnerable to its weaknesses, and so is your audience. By searching within specialized databases or sites, you’re less dependent on a particular search engine.