Comments on: Lookery Site of the Week http://www.feld.com/wp/archives/2008/11/lookery-site-of-the-week.html Mon, 22 Mar 2010 12:23:09 +0000 http://wordpress.org/?v=2.9.2 hourly 1 By: Brett http://www.feld.com/wp/archives/2008/11/lookery-site-of-the-week.html/comment-page-1#comment-10088 Brett Wed, 07 Jan 2009 01:55:58 +0000 http://www.feld.com/wp/archives/2008/11/lookery-site-of-the-week.html#comment-10088 Every visitor for the month of November is 25 to 34? Every visitor for the month of November is 25 to 34?

]]> By: Scott Rafer http://www.feld.com/wp/archives/2008/11/lookery-site-of-the-week.html/comment-page-1#comment-9821 Scott Rafer Wed, 07 Jan 2009 01:55:57 +0000 http://www.feld.com/wp/archives/2008/11/lookery-site-of-the-week.html#comment-9821 The issue gets an 80-20 kind of answer from Lookery's end. The 80% is that we're only collecting and displaying empirical data. We don't attempt to extrapolate or otherwise estimate information about your audience that we don't know for certain. Our early data sources skew towards that age group, so it's over-represented for now. The 20% is that between now and early December we're moving over to a new data crunching backend. It scales more reliably than our current one, plus it's far more coolio (to channel Marcoullier for a moment) -- all Hadoop and Erlang on AWS as opposed to other, lesser stuff all on AWS. I wouldn't be entirely shocked if our current backend was dropping a few data points on the reporting side (as its priority is our commercial API). The new backend is supporting keyword analysis already. Age and Gender will be phased in progressively, hopefully so no one notices anything but gentle improvement. The issue gets an 80-20 kind of answer from Lookery's end. The 80% is that we're only collecting and displaying empirical data. We don't attempt to extrapolate or otherwise estimate information about your audience that we don't know for certain. Our early data sources skew towards that age group, so it's over-represented for now. The 20% is that between now and early December we're moving over to a new data crunching backend. It scales more reliably than our current one, plus it's far more coolio (to channel Marcoullier for a moment) — all Hadoop and Erlang on AWS as opposed to other, lesser stuff all on AWS.

I wouldn't be entirely shocked if our current backend was dropping a few data points on the reporting side (as its priority is our commercial API). The new backend is supporting keyword analysis already. Age and Gender will be phased in progressively, hopefully so no one notices anything but gentle improvement.

]]> By: Rex Dixon http://www.feld.com/wp/archives/2008/11/lookery-site-of-the-week.html/comment-page-1#comment-9825 Rex Dixon Wed, 07 Jan 2009 01:55:57 +0000 http://www.feld.com/wp/archives/2008/11/lookery-site-of-the-week.html#comment-9825 If you ever have any questions, please feel free to drop me a line directly. I hope Scott's explanation above has helped you (or anyone using our services) understand. The great news is that our new very fast number crunching system is almost live. Our devs in one word, ROCK. :) If you ever have any questions, please feel free to drop me a line directly. I hope Scott's explanation above has helped you (or anyone using our services) understand. The great news is that our new very fast number crunching system is almost live. Our devs in one word, ROCK. :)

]]> By: bfeld http://www.feld.com/wp/archives/2008/11/lookery-site-of-the-week.html/comment-page-1#comment-9845 bfeld Wed, 07 Jan 2009 01:55:57 +0000 http://www.feld.com/wp/archives/2008/11/lookery-site-of-the-week.html#comment-9845 I guess I attract a very narrow demo (oops). I saw that also – I’ll check in with Lookery since that clearly isn’t correct.<br /> I guess I attract a very narrow demo (oops). I saw that also – I’ll check in with Lookery since that clearly isn’t correct.

]]>