enjay Posted January 20, 2009 Posted January 20, 2009 We've been asked to produce traffic analysis for our web site - just the basic stuff like number of visitors, most common pages, etc. Our ISP doesn't offer anything like this, instead just sending us access log files and telling us to use some analysis software to read them. So - what are people's preferred tools for reading these files? A quick (and possibly poor) Google search came up with too much noise to be useful...
WithoutMotive Posted January 20, 2009 Posted January 20, 2009 Another vote for Google Analytics here.
enjay Posted January 20, 2009 Author Posted January 20, 2009 Is that looking at our own log files, or would we need to put some code on the pages to send the traffic data to Google? I think that I'm more after an application which can present the data we already have, rather than a service gathering the data over again, although I'm not entirely sure why I prefer that!
tom_newton Posted January 20, 2009 Posted January 20, 2009 Try the top couple on this list... webalizer certainly works ok freshmeat.net: Browse project tree - Topic :: Internet :: Log Analysis
WithoutMotive Posted January 20, 2009 Posted January 20, 2009 Is that looking at our own log files, or would we need to put some code on the pages to send the traffic data to Google? You place code on the pages and Google track the data and present reports to you. You can choose to have them emailed to you, too, monthly or even weekly. There's lots of information that it presents to you.
DanW Posted January 20, 2009 Posted January 20, 2009 Mint: A Fresh Look at your Site verry nice, simple, fast, cheap and lots of plugins 1
Martin Posted January 20, 2009 Posted January 20, 2009 The trouble with Google Analytics is that if you are the slightest bit web-savy (or know someone who is) you can install NoScript (under Firefox) and specifically block any Google Analytics scripts. Thus you cannot trust the reports that it produces, no matter how "impressive" they may be. As a user, you may also want to think about the implications of having Google analyse data extracted from web-pages, YouTube, Google Earth, GMail etc. They can start to build up quite a profile of you as not only do they know what you look at, but know who you e-mail and probably know where you live ('cos we have all Google-Earthed our own house haven't we ). mb
matt40k Posted January 20, 2009 Posted January 20, 2009 Another vote for Google Stats, also Awstats. Google stats add a javascript, simple c'n'p to the bottom of your webpages. Awstats looks at your logs. I like both, there are some difference, also the figures will be different, if the end-user blocks javascript. Also awstats should be able to view your previous (pre-installation) logs
enjay Posted January 20, 2009 Author Posted January 20, 2009 We're going to try Absolute Log Analyzer - flexible web site statistics - only $50, and since it looks at our log files, NoScript etc won't block it.
enjay Posted January 20, 2009 Author Posted January 20, 2009 (edited) As a user, you may also want to think about the implications of having Google analyse data extracted from web-pages, YouTube, Google Earth, GMail etc. They can start to build up quite a profile of you as not only do they know what you look at, but know who you e-mail and probably know where you live ('cos we have all Google-Earthed our own house haven't we ). Maybe I'm missing something, but so what?! I can't say I'm losing sleep knowing that Google have the potential to know my postcode and who I email. This is nothing new anyway - Facebook know all sorts of stuff about me, and Amazon know where I live, where I work, what I read, watch and listen to! Again, none of that concerns me. Ultimately, what these companies can achieve by assimilating this data is presumably a better, more targeted service for me the customer. Result, I say. Sadly though, they're not using it, so Amazon continue to spam me about the latest Girls Aloud CD or whatever, when they could easily work out that I have no interest whatsoever in it. Edited January 20, 2009 by enjay
Zimmer Posted January 20, 2009 Posted January 20, 2009 Webalizer is indeed very good for Apache web statistics. We also use use ISA log files and sometimes data from Google Webmaster Tools.
zag Posted January 21, 2009 Posted January 21, 2009 We use google analytics for online stats. Google Analytics http://creativebits.org/files/google_analytics.png and Web Log Expert for our local IIS logs. WebLog Expert - Powerful log analyzer http://www.weblogexpert.com/images/reportshot_s.gif
srochford Posted January 21, 2009 Posted January 21, 2009 We've been asked to produce traffic analysis for our web site - just the basic stuff like number of visitors, most common pages, etc. Our ISP doesn't offer anything like this, instead just sending us access log files and telling us to use some analysis software to read them. So - what are people's preferred tools for reading these files? A quick (and possibly poor) Google search came up with too much noise to be useful... Microsoft log parser can read most web log files (not just IIS but Apache and anything else which logs in W3C log format). You can query directly against the text files (or, probably better) import the data to a SQL database and then do your own queries. There's some good stuff here about how to use log parser.
danrhodes Posted January 21, 2009 Posted January 21, 2009 I use Trace Watch on all my websites (50+) its brilliant and works really well. D
enjay Posted January 21, 2009 Author Posted January 21, 2009 As I said, I'm more after a log file reader than script-based online things like Google Analytics. As an aside, it is amusingly difficult to find Google's page about Analytics - I Googled "Analytics" and "Google Analytics" without success before coming back to this page and using someone's link to it!!
Geoff Posted January 21, 2009 Posted January 21, 2009 (edited) If you want to do this yourself with your web server logs have a look at Analog + Report Magic. Analog: WWW logfile analysis Report Magic http://www.chiark.greenend.org.uk/~sret1/stats/ http://www.chiark.greenend.org.uk/~sret1/stats/reportmagic/ Edited January 21, 2009 by Geoff
Martin Posted January 21, 2009 Posted January 21, 2009 Maybe I'm missing something, but so what?! I can't say I'm losing sleep knowing that Google have the potential to know my postcode and who I email. This is nothing new anyway - Facebook know all sorts of stuff about me, and Amazon know where I live, where I work, what I read, watch and listen to! Again, none of that concerns me. Ultimately, what these companies can achieve by assimilating this data is presumably a better, more targeted service for me the customer. Result, I say. Sadly though, they're not using it, so Amazon continue to spam me about the latest Girls Aloud CD or whatever, when they could easily work out that I have no interest whatsoever in it. People post all kinds of personal stuff about themselves on Facebook (sometimes with terrible consequences), but you choose whether you participate or not. Similarly with Amazon, you can choose whether you use them. However Google (and Analytics) hide behind a great number of "innocent" websites and can cross-reference an awful lot of data - and data protection laws in America aren't quite the same as in the UK. As data mining increases, and people become more aware of the consequences, they will start to take action to limit their exposure - things like NoScript, Popup-blockers etc. Thus the data from Google Analytics becomes unreliable (trying to bring it back to topic )! Watch Pizza Palace for an example of how things might turn out! mb
enjay Posted January 21, 2009 Author Posted January 21, 2009 Again, I'm still not really seeing the issue with Google having the potential to compile a list of all the sites which my IP address has visited (and I stress potential, because let's face facts, they're not actually doing this). Assuming everything which I am doing is legal, what concern should I have that Google can see I shop on Amazon and also read KittensMonthly.com?
enjay Posted January 21, 2009 Author Posted January 21, 2009 Okay, fell at the first hurdle here - our website uses PHP, so page addresses are things like domain/index.php?id=7 or domain/index.php?id=34 and so on. The analyser which I had downloaded sees all this traffic as being to the same page, which of course it isn't. Looking at the log files in Notepad (ugh!), the files do contain the ?id=23, so it is obviously just the analysis application is ignoring this part. Of the applications mentioned above (again, at this stage I'm not interested in web-based services), does anyone know which ones can make the distinction between such pages?
enjay Posted March 25, 2009 Author Posted March 25, 2009 Sorry to dredge this one up again, but I'm still struggling to find a log file analyser which can determine exactly which page is being visited. Our site uses PHP, so page addresses are of the format http://domain/index.php?id=2. All the analysers I've found report hundreds of visitors to index.php without breaking down which page they actually visited. The reader needs to install onto Windows and be fairly straight-forward to use by a restricted (staff) user, i.e. no visibility of C: drive, no command prompt, etc. Any suggestions which people can offer would be greatly appreciated.
webman Posted March 25, 2009 Posted March 25, 2009 Don't think this was mentioned before, so try BBClone.
enjay Posted March 25, 2009 Author Posted March 25, 2009 Unless I'm missing something, BBClone installs on the web server itself not on a client PC here. Correct?
webman Posted March 25, 2009 Posted March 25, 2009 Unless I'm missing something, BBClone installs on the web server itself not on a client PC here. Correct? Correct, but it's PHP and you can browse to the analysis page from any computer...
Recommended Posts
Create an account or sign in to comment
You need to be a member in order to leave a comment
Create an account
Sign up for a new account in our community. It's easy!
Register a new accountSign in
Already have an account? Sign in here.
Sign In Now