Cuil: A Brief Review

In a day when the term “google” has become synonymous with internet search (in fact, it’s even in the dictionary), a new search engine named Cuil (pronounced “cool”) is making some pretty bold claims.

Cuil searches more pages on the Web than anyone else—three times as many as Google and ten times as many as Microsoft. [source]

Coming out from under wraps just recently, Cuil is attempting to redefine the way web search is done. After noticing it via a news story on the home page of CNN.com today, I decided to take Cuil for a spin to see what it is made of. Some of what I found was nice, some wasn’t impressive at all. Let’s start with the good news.

Good News

Probably the first thing that strikes you is the fresh design and layout. I don’t think many people readily admit that Google has the prettiest site design out there. In fact, it’s pretty bland. But hey, when you’ve got traffic like they do, I suppose something must be done to keep bandwidth in check.

Speaking of the design, I’ve heard that a black Google homepage would supposedly be good for the environment… maybe Cuil’s got something going there.

I also kinda like everything that Cuil showcases on their Features page. Category drilldowns and tabs would certainly help when searching for terms with several possible meanings. Navigation Suggestions are always nice. I’ve personally been taking advantage of Google Suggest for some time (in my Firefox search toolbar).

The column approach to displaying data is certainly a good way to cut down on wasted page space that you get when viewing Google with a larger resolution monitor.

In addition, including relevant images with the search results certainly makes it easier for you to scan the page and find a helpful result faster.

Bad News

I’m sorry to say that searching for info on Cuil seems to be leagues behind Google’s super-refined algorithms. To test out the differences between these two search engines I tried some identical searches on each of them to see what results were handed to me. First, I tried finding my web design company, xHTMLpro.com using some appropriate keywords that have worked well in the past.

As a side note, while I was preparing this blog post, Cuil’s servers bent underneath the load of traffic and consequently were not returning any results whatsoever. That’s never happened with Google and it certainly highlights Cuil’s unreliability as a personal search engine replacement.

Due to excessive load, our servers didn’t return results. Please try your search again.

1. “web design lebanon, pa” (Google, Cuil)

• Google immediately recognized my attempt to search for local businesses and returned a listing of several design firms in my city. Look. There I am … letter H.
• Cuil came up empty handed! I had to try a few more queries before I even got results. “web design lebanon pennsylvania” returned a bunch of links about tea… [scratch head] Google gave me the same results for that string, too.

2. “xhtmlpro” (Google, Cuil) – I decided to try an “easy” query on both

• Google: bam. At the top where I ought to be.
• Cuil: These results are actually relevant… but what is that image next to my site’s listing? I have no idea where that came from.

Ok, let’s try searching for my own name.

3. “peter anglea” (Google, Cuil)

• Google successfully returned several relevant results on my name. The first 5 results concern me, and a few others as well.
• While Cuil actually returned several results relevant to my name, they didn’t even include my home page among them. Instead there are a few links to other pages on my site.

So, needless to say, Cuil hasn’t proved that their method of searching the web is better yet. I concede that it is new and it probably needs some time to learn from its users so that it can return better results. Right now, it’s kinda lousy actually.

But wait a minute… I’m not sure how Cuil plans to improve their results (and they sure need to) since they state they don’t follow click-throughs or collect any user info.

Because Cuil analyzes Web pages and not click-throughs, we don’t need to know your search history and habits. So our privacy policy is very simple: when you search with Cuil, we do not collect any personally identifiable information, period. We have no idea who sends queries: not by name, not by IP address, and not by cookie. Your search history is your business, not ours. We don’t need to keep logs of our users’ search activity, so we don’t. [source]

That’s part of what makes Google’s results so refined. By collecting data about my search habits (by means of being logged into my Google account) they can give me results that are personalized and ultimately more relevant. I agree that privacy is important on the web, but I’m comfortable with Google’s privacy policies, and hey, they give me what I want almost every time.

Going back to the columns idea … while the column approach could an effective way of displaying data, I’m not sure the world is used to that. Every other search engine in the world lists their results in a vertical list. I for one am not used to scanning search results from side to side to find what I want. Does it actually save time? I find that I’m spending more time looking around, and that delays me finding the links I really want.

Also, it’s nice that Cuil is providing additional helps for its users (Tabs, Drilldowns), but that approach can also result in taking more time to find what you want. Your average internet searcher just wants to see what they want. Right away. While Google is constantly evaluating their user experience so that you can shave split-seconds on your search time, Cuil must think that the better way includes giving users more links to follow.

Much more could be said I suppose. I won’t even touch on the claim that their index is 3x larger than Google (Google did just clain they scan over 1 trillion links now). Feel free to comment on your experience using Cuil (pros, cons, indifferences). As for me, I think it will be some time before they’re actually ready to compete with Google. Right now, it’s just not “cuil.”