Friday 9 November 2007

How Facebook Will Break Into Search

Earlier this year Sproose was launched, a new kid on the block in search.

Using the principle of combining regular algorithmic search engine results with "votes" from users, it's kind of a cross between social bookmarking (think del.icio.us / Digg / Reddit) and the established search engines.

Why is this significant?

Search engines depend on relevancy. If users say "yes this is great site", it probably is. If you can capture that democracy and add it to a wider crawled base of sites, you end up with potentially relevant results.

Does it work?

Not yet. This project really depends on getting critical mass. I looked at the high volume terms for the travel sector "cheap flights" and "hotels" and there were 1 or 2 votes for the top 10 sites listed. These terms get hundreds of thousands of searches a month yet there's only 1 or 2 votes for the top ranked sites.

So if Sproose doesn't work, what will work?

If you combine an established bookmarking site such as Facebook, with an established search engine such as MSN, you could do the same thing, and (most importantly) get it in front of millions of users who would adopt it. If you had an advertising network to monetise that traffic, you'd be onto a winner.

Now didn't Microsoft just buy a share in Facebook?

No comments: