Cuil, the latest search engine startup to come out swinging from its corner with the hope of knocking out Google, is instead taking a beating that could do it long-term damage as a credible contender.
The company received broad media coverage on Monday, primarily because it has former Google engineers on its team and because of its claim to have the world’s largest search index, but Cuil is now facing an angry backlash.
The site had performance and availability problems throughout launch day, and a growing chorus of search market observers has declared the engine’s results to its queries unimpressive.
In addition, the site was launched with only a Web search engine, at a time when it’s considered a basic requirement for any search engine with aspirations of competing against Google to have at least some basic vertical search tabs for news articles, photos, maps and local business information.
Without the ability to collate general search results with photos, news articles, the increasingly popular video files and mapping information, a search engine is effectively unable to provide the type of “universal” search results that have become de riguer and that Google now consistently does.
In short, what looked like a successfully architected public-relations launch is backfiring because the product has failed to meet the lofty expectations that were created.
Backed by reputable investors that have reportedly invested $33 million in it, Cuil will now have to go into damage control instead of riding the early momentum, and hope that end-users and industry observers will give it another chance once it works out its kinks.
“First impressions count a lot,” said industry analyst Greg Sterling of Sterling Market Intelligence, who on Monday morning was unable to give Cuil a thorough test because of the site’s performance problems.
While Cuil could become an interesting alternative to Google and the other major search engines, that will take some time. In hindsight, it might have been wiser for Cuil’s management to launch the site more quietly and give themselves time to improve it, Sterling said.
Instead, by making a big, noisy media splash, they gave themselves little room for error. “This much media coverage creates high expectations,” he said.
Indeed, Cuil came out with an in-your-face attitude, claiming to have the largest Web index of any search engine: 120 billion Web pages, which Cuil states is “three times more than any other search engine.” It was a point that its officials made in interviews with media outlets prior to Monday’s debut.
But the site seemed unable to handle some softball, straight-forward queries. For example, a search for “barack obama” returned on the first page of results mostly links to different pages of Obama’s official site, hardly useful if one is looking for a variety of sites on the presidential candidate. A search for “St. Louis, MO” initially returned zero results.
Among others, search engine expert Danny Sullivan criticized Cuil for focusing on the size of its index—a practice that fell into disfavor years ago, as engines focused on the quality of results—and questioned the basic validity of the claim.
“Yes, size matters. You want to have a comprehensive collection of documents from across the web. But having a lot of documents doesn’t mean you are most relevant,” Sullivan wrote on his Search Engine Land blog.
Sullivan also pointed out that Google hasn’t publicly stated the size of its Web index in years, and that even if Cuil’s is indeed three times as big, Google could quickly match that by simply becoming a bit less selective. Last Friday, Google, likely anticipating Cuil’s launch on Monday, said its crawlers today “see” more than 1 trillion URLs on the Web.
Google declined to comment about Cuil’s claim and also declined to say how many links are in its Web index.
Sullivan and other search market observers said they were underwhelmed by the quality of Cuil’s results.
“With the huge caveat that nine queries are far from letting anyone conclude anything, I still didn’t come away with a sense that Cuil has Google-beating relevancy. Instead, it has some flaws though is better than many start-up search engines appear out of the box,” Sullivan wrote in another post Monday.
“I played with the site a fair bit when it turned on this morning. So far it doesn’t do much for me,” wrote Saul Hansell of The New York Times in a blog post titled “Cuil’s New Search Engine: Cheaper Than Google, but Not Better.”
Cuil, which is pronounced “cool,” got a resounding thumbs-down from The Wall Street Journal’s John Paczkowski in an AllThingsD post titled Totally UnCuil.
“If your mission is to beat Google in the search market, it’s probably wise to give your upstart search engine a name that people know how to pronounce. It’s also wise to make sure that it appears in the first page of search results for its own name. Cuil, the upstart search engine that debuted today with aspirations of unseating Google, has apparently done neither,” he wrote.
He’s not alone in his dislike for the company’s name. IDC analyst Caroline Dangson flagged the name—she called it “terrible”—as one of what she considers the company’s main challenges.
“Cuil has an uphill battle in getting more consumers to search its site instead of Google. Google wins hands down for brand recognition among U.S. consumers,” she said via e-mail, adding that using Google has become a habit for a majority of search users.
According to a recent IDC survey, three-quarters of U.S. online consumers know Google as an Internet brand and the majority like Google and believe it offers quality services, she said.
“Furthermore, Google also already has an established business model based on search advertising in which it excels. Even if Cuil offers better search results, the company is not monetizing its service with advertising at this point and will not be able to compete with Google in terms of revenue,” she said.
Of course, on Monday, Cuil’s problems were much more serious than offering better results, considering how shaky its performance has been. At close to 5 p.m. Eastern Time the site was still returning error results, saying it couldn’t process queries because its servers were overloaded.
“From the perspective of Web search technology, Cuil has a number of hurdles to overcome, as its failure to survive a day in the sun today proves,” said IDC analyst Susan Feldman.
The performance problems at Cuil are not likely due to a lack of technical expertise, but most probably an issue of limited infrastructure resources, Feldman said in an e-mail interview.
“Scaling up to billions of queries and documents is a problem in and of itself. That is probably the main hurdle that most startups have to jump. Doing this day and night, and keeping the index current while providing sub-second responses takes a toll on the performance of any system,” she said. “The problem is that this requires server farms that are usually outside the scope of a start up, and therefore has constituted a barrier to entry.”