--
You received this message because you are subscribed to the Google Groups "Urchin 6" group.
To post to this group, send email to urchin...@googlegroups.com.
To unsubscribe from this group, send email to urchin-help-...@googlegroups.com.
For more options, visit this group at http://groups.google.com/group/urchin-help-6?hl=en.
Overall it's sad to see Urchin being retired. It really is a great product and there's very little for alternatives out there. We've been reviewing other solutions and are coming up pretty empty handed so far. Systems like Splunk that purport to be great for Web Analytics are really not up to the job compared to Urchin. WebTrends is largely out of reach for most Urchin users as a solution, and the open source and free options are absolutely terrible. So... keep running Urchin? :)
Agreed. I presume part of Google's motivation for eliminating Urchin is to drive more customers to Google Analytics, Premium or otherwise. Given the difference in cost I don't know if many customers will make that jump to Premium.
For our needs, we have to consider moving to Premium. We started on Google Analytics, then begin migrating to Urchin two years ago, and finished that transition last year at the same time we upgraded to 7.0. So needless to say, the timing was horrible for us. We plan to run Urchin 7.2 at least through the end of 2012, but will start planning a migration in the meantime.
In our production Urchin server today we have over 900 profiles and hundreds of users, and handle millions of visits each day. The product has a few bugs we worked around (e.g. parallel processing doesn't work for remote log sources) but overall it's done a great job. We were able to automate the management of profile for the most part, thanks to an intuitive MySQL schema for profile/filter/log/user metadata. That was one of our pain points with GA, requiring tedious management of accounts and profiles. We also use the data API extensively—we can drive the Urchin API hard, extracting tens of millions of records a day if needed, but GA restricts our API use with CAPTCHA challenges.
It's not entirely clear whether GA Premium would solve all our challenges, and I'd imagine we'll be speaking to a reseller soon to learn more and/or obtain a trial edition.
If anyone who still reads this forum discovers a successful migration path off Urchin, regardless of the product/vendor they choose, I'd love to hear about it. Would Google mind if we used this forum to discuss migration? I don't see why they would, they're abandoning us anyway :-(
-Jeff
You surely know from reading the documentation that GA does not collect personally-identifiable information (provided you don't supply PII in your tracking code, via e.g. custom vars). But if your intent is to prove this because you don't trust Google's claims, you'll need to find a testing approach that is suitable to a PCI auditor.
Maybe your testing efforts should focus on the actual tracking data. Tools like Firebug can capture the tracking pixel data transmitted to Google, and you could analyze their contents. Provided the tracking data isn't overly obfuscated, it should be straightforward to demonstrate no match for personal data. You'd need to do this testing on the browser, because attempt to intercept and read the packets at a network level will be thwarted by the SSL encrypted payload.
If, on the other hand, you don't trust that Google will never change their tracking to begin collecting such data (either knowingly or because they have been compromised), I don't have an answer for you. I'm confident Google won't allow this, because if discovered it would effectively put them out of business, but I have no idea if that explanation would satisfy a PCI auditor. My feeling is that Google's privacy policy combined with reasonable testing should be adequate.
-Jeff