Web Analytics with Treasure Data – Ask the Tough Questions
Like you probably do, I use Google Analytics for basic web analytics. It provides a quick view into how traffic is growing on www.treasuredata.com and several other domains we own. It also provides an easy view into the paths that our visitors as a whole typically follows through the site.
Identifying the sustained popularity of Google Analytics, competitors have come from all sides. If you are interested in funnel or cohort analysis, Mixpanel seems to be the tool of choice these days. If you need to integrate with a more comprehensive marketing suite, Omniture has evolved into a component Adobe Marketing Cloud since it was acquired in 2009. Every year, it seems like a new tool is becoming the dashboard of choice for progressive web marketers.
But what happens when you need to go beyond dashboards for web analytics. What happens when you want to dive into the raw row-level granular data behind the pretty pictures? Here, we eat our own dogfood at Treasure Data.
Now, our web analytics are no longer limited by the questions someone designing a dashboard thought I would want to ask. Don’t get me wrong, some of those folks have done great jobs, but as with most other types of analyses, web analytics often makes you peel off more layers of the onion than you initially expected. With SQL access to our raw web data, I can now ask questions like “Show me all the visitor IDs who hit the trial signup page and pricing page, but didn’t sign up. OK. Now, show me all the visitors who hit the pricing page but did not hit the homepage or About page.”
If I want to get more complex, I can easily join data across sources – Marketo or Salesforce.com, for example – with our web analytics.