PlayerSearch Blog

by Ted Kasten, Founder of PlayerSearch

Sports enthusiasts don’t use Google to find Sports content

According to the most recent Hitwise press release, only 12.93% of visitors to sports related websites arrive through the top 50 search engines. This is the lowest figure of any content category including Health, Travel, Shopping, Entertainment, and Finance. Combine this information with the fact that demand for sports content is higher than ever and I think it is clear users know that general search engines do not provide very good results for sports searches. Most sports enthusiasts go directly to ESPN and the local website or blogs for their favorite team. They understand there is a lot more news coverage available, but they don’t have the time to go to each individual site, or find new sites, to search for their coverage of their favorite players or teams (sports fans demand for sports news and analysis is unlimited…but their time is not). At this point all sports fans are well aware of Google, but they clearly don’t go there either because the search results are not very good for sports.

I believe this is a clear indication that there is a huge void in the sports search space with incredibly high demand for sports content (over 60 million people visiting sports content sites and about 15 million people playing fantasy sports games) yet no great search solution. Clearly people are not going to Google to get their fix of sports news and analysis. The top search engines currently account for 45.10% of the traffic to health related sites (this figure includes traffic from the top general search engines such as Google as well as health specific search engines like Healthline; I don’t have market share data but clearly Healthline has been able to capture a small but significant share of health related searches). If we can get the search experience for sports correct, there is clearly a lot of room to grow this category.


April 8, 2008 Posted by | Google | , | Leave a comment

Mahalo Demo by Jason Calacanis at All Things Digital Conference

Jason provides a quick demo of Mahalo at the All Things Digital Conference. It is interesting to hear him discuss his approach to improving search by making it human-powered. I think what Mahalo is doing is great and it will be very interesting to see how 100% human-powered search scales and if it can be profitable. Mahalo has apparently raised 5 years worth of cash, so they have a long runway to make this work.

Some of the key aspects to his model include a more friendly looking User Interface and “potentially” better search results/content than is provided by algorithmic search. Jason’s primary complaint with algorithmic search is that an entire industry has been created with the sole purpose to game these algorithms for financial gain and at the searchers expense – very true. So the potential for human “Guides” to manually create search results page takes SEO spammers out of the game and can theoretically provide superior search results. Mahalo is going after the top 10,000 search terms which Jason states covers 24% of all English language searches while leaving the remainder of the long-tail searches to Google.

There are three categories of search – navigational, informational and transactional. I see Mahalo having the greatest advantage with transactional searches as these searches get spammed the most and a human can better evaluate hotels, cars, computers, etc. for categories such as quality, price or location. Informational searches (which is what PlayerSearch focuses on for sports information) can also benefit from cleaner results created by human editors or guides. I doubt Mahalo cares about navigational search as there is typically no revenue in these searches (they let Google handle these).

One challenge everyone points to for Mahalo is their ability to scale and maintain high quality, fresh search results. Jason states that Mahalo creates 500 pages a day with a staff of 40 people (growing to 100 people) so one page takes about an hour to create. The real challenge is in maintaining the news at a level that people will keep coming back for. To keep the news fresh they supplement their own content with links to other popular sites (a search on Mahalo for Brett Favre has a link to the Brett Favre results on PlayerSearch!).

Mahalo’s approach to improving the overall search experience is similar to that of PlayerSearch. We both improve search results by replacing algorithms with human editors and by categorizing the content logically. We also improve the user interface and provide blended results. Mahalo is taking on a much larger opportunity while we are focusing only on sports. I think the content included in the PlayerSearch results is an excellent supplement to Mahalo’s search results (see the Brett Favre link above).

Mahalo’s initial press release is here.

April 8, 2008 Posted by | Uncategorized | Leave a comment

Searchers prefer blended search results

Bill Morrison at ThinkPanmure (formerly ThinkEquity) just forwarded me an interesting article on users preference for blended search over vertical search (image, news, video) by (study conducted by Jupiter Research). In this case vertical search refers to images, news, and video, not sports, finance, health, travel, etc. (so this supports PlayerSearch’s approach of using blended search results that currently include sports news, videos, photos and stats; we do not force users to search by one of those categories although they can click the View More link to dive deeper into most of the content categories).

Here is one of the charts:

Vertical search not prevantly used on major search engines

The article outlines some interesting stats that point to users preferring blended search results that include images, news and video directly in the search results page as opposed to having to click links to get only news or only images or only videos. This finding is well known, but it is always good to find hard numbers outlining users behavior.

April 8, 2008 Posted by | Google, Universal Search | , | Leave a comment