We're less than 24 hours removed from Facebook's big announcement of Graph Search, and despite all the laudatory praise for the idea of the thing, early reviews have pointed to one very big issue: Graph Search doesn't always work. It's not glitchy, per se; it handles natural language well, says The Verge's Thomas Houston, and people seem to like the "amazing user interface and flexibility," too. But the "third pillar" of Facebook just doesn't yet have enough data to offer relevant answers to social-search questions. The results are "lacking" says Engadget, and Gizmodo calls this "hollow" experience an "authority problem." Or more bluntly, as writes Slate's Farhad Manjoo put is, "The search results aren't that good." That's right, for all the users and user information the social network has compiled in its nine-year history, Facebook search has a data problem. So, despite looking like a "visually rich beast" with the potential to kill Google, Yelp, OkCupid, and LinkedIn, Graph Search doesn't sound like a replacement for any of those services because users haven't handed over quite enough intel for all the right answers — at least not in the first 24 hours.
Specifically, Graph Search's early beta users and reviewers found that the proprietary Facebook algorithm didn't offer useful results. For example, when Manjoo searched "Restaurants liked by people who live in Palo Alto, California," Facebook recommended the Facebook Culinary Team. "Now, I’ve had a couple good sandwiches at Facebook’s café, but I don’t remember ever being knocked off my feet," he explains. After a series of successful searches, SearchEngineLand's Danny Sullivan encountered a similar problem while searching for restaurants in Newport Beach: "Trust me, Newport Dunes isn't a hot restaurant."