Hi I am in the process of setting up a unique social network (not another myspace/facebook clone failure dont worry). The problem I am having is trying to predict how many page impressions the site will get in relation to how many users are online at once. I own a few sites including a forum, but I can't seem to work out how many users online will affect page impressions on a social network. This site: tinyurl.com/4pfnyc has about 800 people online at once. I would like to use this as a benchmark for my site. Could anyone make any justified predictions to how many uniques/page impressions this site gets per day with 800 people online at once? Any help would be much appriciated as this will help me predict ad revenue. Thanks
FYI,page per impression is cannot be accurated predicted coz it work like random behavior. But you can predict it if only you have the history data of it. Page impression is not based on how many people online but it muchly depend on how many page those people surf the pages of your site.
@Aciid - I created an Excel model to estimate a website's traffic based on Alexa's 3 month reach average. You can read the post in my blog and even download the Excel file that I have there. http://mbainternetmarketingmanager.blogspot.com/2008/10/website-traffic-estimation-my-excel.html I hope it's useful for what you're looking for
http://www.cubestat.com/www.musicv2.com Seems 800 online might be a bit exaggerated, as quantcast + compete are more accurate than Alexa and they indicate, like Alexa that this site does not have a lot of traffic at all..
Well, I think your bandwidth would be pretty big and you would need a dedicated server for sure if you want to expand anywhere beyond that.