Google Analytics New Features: Intelligence Engine, Custom Alerts & More
Google has just announced a number of new features for Google Analytics, including more powerful reporting capabilities, greater customization options and a new “intelligence engine” that Google says can help search marketers drive smarter data insights. Here’s a rundown of the new features, and why they’re important.
Google Analytics new “intelligence engine” comes with default alerts that will show you interesting trends in the data: it surfaces insights. Exactly how the algorithm that does this works is not clear, but it certainly takes into consideration many variables, in a very Googly manner. This can be a great way to start your day looking for insights—a kind of a conversation initiator for web analysts.
As we can see in the screenshot, Google did not let us down when it comes to design and UI. The interface is very intuitive and lets the analyst move through new alerts quickly.
Some really interesting things you can do:
Define your alert sensitivity. This controls the number of automatic alerts that will be triggered to appear in the Intelligence report page. As Google explains, “a higher significance rating is assigned as the difference between the actual performance increases with the expected performance.”
Create advanced segments based on alerts. On the right column of each alert you will see a link to create a segment. This is highly useful, since through alerts you might discover high converting segments that should be tracked separately, so Google makes this segment only one click away.
Creating custom alerts
Creating alerts is a very effective way to track visitor segments that are specially interesting to you. Let’s say you run a banner campaign to bring people into the website. How can you decide when to stop the campaign? Now you can use alerts to have an email delivered to you once your campaign bounce rate increases by 10% compared to the same day in the previous week. This is valuable because the website visitors (where the banner is shown) might get used to it, meaning that you reached those users that were high probability prospects. The option to compare to the same day in the previous week is a very good addition since it accounts for weekly variability.
For help on how to create custom report you can visit this help article
Other new features
In addition to the intelligence engine and the ability to create custom alerts, Google has also enhanced Google Analytics with the following new features, according to this Google Analytics blog post:
Expanded goals & site engagement goals. Two new goal types allow you to measure user engagement and branding success on your site. The new goal types allow you to set thresholds for Time on Site and Pages per Visit. Furthermore, you can now define up to 20 goals per profile.
Advanced analysis features (incl. advanced table filtering). Google Analytics provides an arsenal of power tools you can use to perform in-depth, on the fly analysis without having to export your data to spreadsheet tools. Using Secondary Dimensions, you can view multiple levels and combinations of data side by side instead of having to drill down into each level. You can then use the Pivoting feature to cross-tabulate two different metrics with two different dimensions.
Multiple custom variables. Custom Variables gives power-users the flexibility to customize Google Analytics tracking to collect the unique site data most important to their business. With this feature, users can classify any number of interactions on the site into trackable segments.
Despite all of these great new features, we will still be dependent on human beings to analyze and take action on the data. Even Google Analytics evangelist Avinash Kaushik has written about his 10/90 rule, which says that just 10% of web analytics relates to tools, while 90% of both cost and success has to do with humans who analyze data and make recommendations for change based on their insights.
Opinions expressed in this article are those of the guest author and not necessarily Search Engine Land. Staff authors are listed here.