Look Out Google Site Search, Lijit Says It's Right On Your Heels
Innovative search startup Lijit has done study of search widgets on pages around the web and says its widget is very close to becoming as popular as Google's own "site search". We think Lijit is quite interesting and we're not surprised to find out that many other bloggers around the web agree.
The Boulder, Colorado company did an interesting survey of the widget-o-sphere, if you will, and found that 45% of the search widgets it found were its own. 47% were from Google. We (more or less) believe these numbers and think we've got some idea why Lijit is as popular as it is.
The Numbers
Lijit defines a widget for the purpose of its survey as: any regularly-occurring functionality on a blog powered by an external service, voluntarily installed by the blog owner, and powered by Flash or Javascript. A simpler definition is that a widget is a little bit of code people drop into their website that makes things happen, either dynamic content or functionality.
Lijit looked at 184,431 blogs and found at least one "widget" on 79.5% of them (146,636 total). 10.39% of those blogs had a widget put on it for search. (We've got search widgets on our site, for example, from our sponsors Quintura and Eurekster.) The company counted more than 1,200,000 widgets in total.
Whose search widgets are most popular? Google's by just a little bit, Lijit says. 47% of the search widgets they found were made by Google services like Google Co-op. Lijit had 45% of the widgets on surveyed pages.
Other popular blog search widgets, trailing far behind, included Sphere, Blogbar.org, IceRocket, Eurekster and Quintura.
Are These Numbers Believable?
We believe these numbers - roughly. We asked competitor Eurekster if they believed that Lijit was nearly as popular as Google site search and they said they were apt to believe it as well. Searches around various social media environments show that people talk about installing Lijit widgets quite often, too.
The company adds the following caveat: "Our crawl is 'centered' on blogs with our Lijit widget. Our crawler then expands outwards by following blogrolls. This will give a bias to the overall results." It sure will! We're not ready to take these numbers to the bank but after swallowing a big grain of salt, we can accept the general conclusion: Lijit search widgets are very popular, popular enough to rival Google search widgets.
Why is Lijit So Popular?
The Lijit user experience is quite good. We've discussed the company's excellent method of getting users to fill out their profiles before. Those profiles enable a blog's visitors to search not just that single blog, but also the author's social bookmarks and other online content. The searches are done without taking readers off-site, results appear in a light-box popup. The results pages are very intuitive, designed to look like the Google Custom Search engine that powers Lijit. Site owners receive a weekly email with interesting analytics about what kinds of searches are bringing readers to their sites.
All of that ads up to a great value proposition. Thus we're not surprised that Lijit is as popular as it is.
Check out this post on the Lijit blog for more details from the company's widget survey, including numbers on which video widgets are most popular, what types of widgets in general are popular, etc. It's interesting data.
Comments
Subscribe to comments for this post OR Subscribe to comments for all ReadWriteWeb posts
I saw this feature on some other sites and I was quite impressed. I don't know if it has ANY chance to take over google, but it is pretty good.
Posted by: Jesse W. | November 10, 2008 5:37 PM
No one will come close to Google in site search or other.
Posted by: venkat | November 10, 2008 8:46 PM
You would think a sponsor's name would be spelled correctly and linked correctly right ?
Isn't it eurekster (eurekster.com) and not eurekester.com ?
And eurekster - if you get to this, add mis spells to your google alerts so you find how people mis-spell you (even partners in this case) since you don't have a name that can be remembered too well.
Posted by: notbornhere.com | November 10, 2008 11:52 PM
thanks..
Posted by: bariyer | November 11, 2008 2:29 AM
I was really surprised how you picked up such a biased research and made a conclusion
Less than 200K blogs were reviewed that represent less 0.1 percent of the blogosphere. In addition, those sites reviewed are mostly from a long-tail of the blogosphere where there is no real money to be made (you know this better than me) since those small blogs get few real visitors (not RSS subscribers) per day each
You can check that out of Technorati Top 100 blogs, more than twenty percent of blogs use Google and only two percent use Lijit (on one, it’s a secondary search engine)
According to its corporate websites, Sphere has more 120K sites using its widget vs. Lijit's 6K, but the latter's research shows that it has almost 10 x bigger market share than Sphere.
How did not you check those facts?
Posted by: Yakov | November 11, 2008 6:32 AM
Lijit is powered by Google's Custom search .. so in turn all the effort lijjit is doing to increase its market share and improving its product is the gain of google at the end of the day. So there's no point of comparing it(lijjit) with google.
Posted by: Supreet
|
November 11, 2008 6:48 AM
Thanks for the nice writeup Marshall. @Supreet, the majority of Lijit's larger publishers use our own search infrastructure rather then Google. Google is a great partner of ours but many of our larger partners need more control and integrations than Google can provide.
We stand by the Widget data in this report. We just report what we see !
Posted by: Todd Vernon | November 11, 2008 6:55 AM
I struggle with this comment "The Lijit user experience is quite good.". Have you actually used the search? it takes ~15 seconds to load, results are no better than google search, people expect instant search results. I've noted several big name bloggers remove it for that exact reason.
Posted by: Ben Young | November 14, 2008 11:54 AM
In addition to user experience, part of allowed Search on blogs should also be the revenue the company is sharing with you, for the revenue driven ones the next step is getting transparency from companies like Lijit on their actual revenue vs. the % the Publisher may or may not get, and use of our content when we decide to use either. P.U.B. has an F.A.Q. on the Lijit issue on our
http://www.lijit.blogspot.com .
Posted by: Barney Moran | November 15, 2008 7:31 PM
@Ben - Im sorry your personal experience with Lijit hasnt been perfect. Was the slow load on your blog, or is that your experience on others blogs? We spend a lot of time on the speed of our service and tracking the speed of searches.
A point of clarification. Generally, we search your blog, content and network and return single results (so three searches for a combined result 3 searches for 3 results) whereas most major search engines do a single search (search their entire index - 1 search for 1 result)
Also, the concept of lijit to search your content and return results centered around your expertise, unlike major search engines which attempt to return results centered on their entire index of data. Its rare that our results and the results from a major search engine match, and its the searcher's perception as to which are better. I, of course, often think we do a better job! :)
If the slow result was on your blog, please email me at micah [at] lijit [dot] com. I would be happy to have our team take a look and see if we can do anything.
Posted by: Micah Baldwin | November 17, 2008 7:53 AM