For a few months I have been hearing the rumblings of a new search engine called Cuil (pronounced cool) that was being started by former Google employees. Well, last night Cuil officially launched to a great deal of buzz and press. The buzz revolves around two key components, first it was started by 3 former senior level Googlers, including Anna Patterson who worked as a senior search architect for Google. The other piece of buzz revolves around the fact that Cuil claims they are the “world’s biggest search engine” – which is a pretty lofty claim no matter how you look at it. On Cuil.com they are currently showing the number of searchable web pages at 121,617,892,992. Impressive…but how real?
Cuil, of course, is not giving away any of its secrets in terms of how it ranks and indexes. However, Cuil did release some insight into their ranking algorithm.
“Rather than rely on superficial popularity metrics, Cuil searches for and ranks pages based on their content and relevance. When we find a page with your keywords, we stay on that page and analyze the rest of its content, its concepts, their inter-relationships and the page’s coherency.
Then we offer you helpful choices and suggestions until you find the page you want and that you know is out there. We believe that analyzing the Web rather than our users is a more useful approach, so we don’t collect data about you and your habits, lest we are tempted to peek. With Cuil, your search history is always private.”
So, what does that mean? Basically, Cuil is focusing on the fact that their relevancy goes much deeper when compared to other search algorithms. But, I am a bit confused when they talk about “superficial popularity metrics”. To me what they describe is what we know about Google’s algorithm. How is what they describe any different than a standard search engine algorithm that builds relevancy through various components like inbound links, on-page attributes, etc? Almost seems like they are talking about a social search site like Digg.
Putting everything else aside, how is Cuil as a search engine? First, lets take a look at examples of the number of indexed pages for both Google and Cuil.
As you can see from the 3 examples above, Google out indexes Cuil on every search. In fact, the above examples are just the 3 I decided to show, there was not a single example of where Cuil returned more results. So where are all of those 120+ Billion web pages? In fact, for the search “homes for sale” Cuil couldn’t come up with a single result while Google had close to 100 million. No search results for a relatively common search term is completely unacceptable.
However, even though common sense says that to have the best search results you need the most indexed pages, the real test comes down to relevance. This is where things becomes subjective. Let’s put aside the indexing numbers for now since a search engine should really be judged by the relevancy of the SERPs – quality over quantity. Unfortunately, Cuil doesn’t beat Google at this either. Whether I searched a broad or specific term Cuil usually displayed some relevant results, but they also had the habit of showing some random and irrelevant results. For example, I keep seeing New Zealand TLDs here and there. Sorry, when looking for “Homes” I am not interested in anything outside of the US. Don’t get me wrong, Cuil’s SERPs are good, however are they better than Googles? No, not from what I see.
Don’t get me wrong, Cuil does have some positives and cool features. I like their layout (3 or 2 columns) and how you can break searches into categories. But, at the end of the day Cuil is no Google or even Yahoo. I see it is a good search engine but it will never compete with the big three. Granted, it is only a few hours old but to compete with Google you need to launch perfectly out of the gate. Cuil could get better over time, but by then who really cares?
Now, in terms of Google killers, I think it is time we start to focus more on “Semantic Search”.
{ 1 trackback }
{ 3 comments… read them below or add one }
I don’t know why this is but if you switch of save search in preferences then you get 22,600,000 results for homes for sale. Maybe Homes is a word for people above the age of 18 :)
Cuil is cool (pun intended). But right now I see to many bugs (Homes being a good example). I agree that you need to go live perfect and so far with Cuil i see many flaws.
It will be interesting to see how Cuil results / interface change over the next few days, weeks and months. I plan to monitor Cuil for a series of keywords and see how the results change over a given time period. Again, I am sure they will improve their results but with all the press they are getting today I am surprised they weren’t better prepared.