In my previous analytics post I discussed the accuracy of web based analytic services. Basically, I cover the inherent flaws in Javascript tools, since most security conscience users will likely disallow Javascript on “untrusted” pages. I recently discovered a new analytics service called Clicky.

Clicky’s service will be a breath of fresh air for someone who has used Google’s service for so long. They offer multiple methods for gathering data, including the standard JS code and a tiny GIF which will allow stats to be gathered even from those who run NoScript or use an offbeat browser.

Detailed list of all of Clicky’s features and see what really sets them apart from Google and others:

I still believe that doing your own analytics is the best way to get accurate data and track errors. For those who don’t host their own blog or cannot use Javascript for whatever reason, I highly recommend Clicky. You may even want to use their GIF based tracking with your current analytics provider by using a <noscript> tag after your JS, just to see what kind of stats you are missing.

Splunk Replay

I am currently comparing the accuracy of Google Analytics against old fashioned log mining using an app called Splunk. I’m starting to get a little fanboyish about Splunk and I’ve even started using it to monitor logs on my own home boxes. Description:

With a variety of flexible input methods you can index logs, configurations, traps and alerts, messages, scripts, and code and performance data from all your applications, servers and network devices. Monitor file systems for scripts and configuration changes, capture archive files, find and tail live application logs, connect to network ports to receive syslog, SNMP and other network-based instrumentation.

The inherent problem with remote Javascript based tools like Google’s is that not everyone uses JS. With the advent of plugins such as NoScript for Firefox, I figured that there would be a considerable margin of error between the two methods. From what I’ve seen so far, I would estimate as much as 20% difference in overall hit numbers. Your results will vary of course. It’s all about your audience.

Though, I must say that these web based tools have their uses. For instance, the typical blogger will likely start their site on a shared host, which generally have either limited or disabled traffic analysis without forcing the customer to pay an additional fee *coughgodaddycough*

For my sites, Splunk gives me what I need most. I want a quick way to view errors, breach attempts, and traffic info. If I feel curious about geographic info, there’s even a Splunk Globe plugin that plots hits on a globe in near real time. As if that’s not enough, some intrepid devs made a replay plugin (screenshot above):

Splunk Replay is a Flash-based, data visualization tool which “replays” your Splunk’d logfile activities in an animated layout. You can click on the image below to see an early version of Replay running on anonymized event data from our internal wiki system.

Splunk

Splunk Globe

Splunk Replay

#splunk on efnet (Amazingly helpful, a breath of fresh air.)

I don’t represent Splunk in any way, they didn’t pay me a dime. Here is the real agenda:
We don’t know a thing about what JS tools like Google’s are doing with your data. They could be selling it to their advertising customers or anyone else for all we know. The point is, if you can do something better, locally, then do it.
© 2010 nukeitdotorg Suffusion WordPress theme by Sayontan Sinha

Powered by CDN Rewrites