Today had a stressful start. I was planning on continuing to update our documentation to reflect our latest features, when our industry partner noticed that our data we collected had some huge discrepancies with the data from their Google Analytics (GA). After hours of troubleshooting, we determined it was not only a problem with GA and Rutilus not being in the same time zone (their GA is configured for PST, Rutilus is configured for UTC, no time zone at all), but the "date picker" GUI component we grabbed for our Dashboard tool was itself using a time zone. It uses whatever the local time zone is, so for us, EST. We finally got it in sync by aligning the query filter to be three hours ahead of GA.

Unfortunately, we still have more discrepancies with the number of visits we're logging than we thought we had. It varies per article, some articles having 10% less visits logged in GA and some having up to 30-40% more visits logged in GA. We're going to try to identify the reason for this with the limited time we have left in the project. Polishing the documentation will take a lower priority while I look into this. Unfortunately, this is a good lesson in why it's important to test as you go when creating large applications like this.


Note: This was originally posted on the blog I used for my co-op term while at Seneca College (mswelke.wordpress.com) before being imported here.