Hacker Newsnew | past | comments | ask | show | jobs | submitlogin
Segmentation of the Top 100 Sites in the US (directededge.com)
28 points by wheels on Sept 1, 2009 | hide | past | favorite | 7 comments


This would be interesting if it aggregated a lot of sites across segments.

Maybe there is only one Music site in the top 100, but the aggregate for Music may well be much higher. That is, what's the aggregate for each segment, what's the average and the stddev.

From this you'd see what segments are more fragmented than others -- i.e. there might only be two "Auction" sites because eBay has it sewn up. There might be only one music site, because there are 1,000s of music sites and nobody's dominated.


To be perfectly honest, your niche is something you perhaps don't want it to be : porn. There is a market, there are customers, and it is where recommendations would be highly valued.

Furthermore, users are willing to rate, there is a lot of tagging going on, the tags are quite detailed, and the benefit from being able to recommend specific items for specific tastes that are not widely found, means that these 'niche' users are easy to monetize.

Your service is perhaps a bit vague at the moment - when you spin off a section that does targeted porn recommendations, you're going to become instant millionares.


Interesting, but not too surprising. How is this going to help you in the recommendations business?


There are a lot of sorts of number that we're running. This was just one of the things that I was curious about today. There's a lot of data available in the US Economic Census breaking down business by size and I wanted to figure out which sectors I might be ignoring in our analysis.

Often when I do stuff like this I like to post it somewhere so that it gets picked up by Google and can save someone else the time later on should it be useful.


Relying on Alexa? Especially in the US Compete.com is the way to go. Alexa is incredibly inaccurate these days.


I don't think compete.com is all that accurate either. I saw some numbers on twitter.com and it reports around 30k uniques per month -- a little low for a site that gets a few million tweets a day. By comparison, quantcast.com reports twitter getting around 31 million uniques per month -- much more plausible.


You probably mistook thousand for millions, I can't imagine Compete.com failing at such a high profile site without them noticing and fixing it.




Guidelines | FAQ | Lists | API | Security | Legal | Apply to YC | Contact

Search: