- Customizable Dashboards
- Emailing of reports
- Improved Segmentation Capabilities
- A general overhaul of the UI
Let's Start at the Beginning Today, I am going to take a bit deeper look at this new release, and I will also look a bit at how it compares to the old release. Let's start by comparing the opening screens.
On the left, we have the old version, and on the right, the new start screen:
Sometimes it's the little things that make the biggest differences. The prior interface looks like it's designed by and for engineers. The new interface simply feels more graphical, and is better looking. I know that's subjective, but I suspect that an awful lot of usability testing went into this release.
The new interface does a much better job of presenting the most common numbers people look for up front. For example, visitors (what some people call unique visitors) and page views are now visible on the initial page. Other key parameters, such as bounce rate, page views per visit, and time on site are also right there.
So right away, we have an opening screen that is a much better fit for senior managers that only care about the most basic numbers. Along those same lines, the ability to email reports was an essential feature for Google Analytics to add. This was one of the largest deficiencies in the product previously.
And it fits into a basic principle of Web analytics: make sure that the person who has to approve the analytics budget (yes, there is still a budget when you use a free tool like Google Analytics, it's for the people who use the tool) feels like they are getting added value from the tools you install.
It's really hard to get a large(r) analytics budget if the people who approve the related expenses don't have a feeling for how it helps the business. Give them something to look at. A simple dashboard that contains the basic numbers they would look for? Absolutely. Have those numbers arrive in their email so they don't even have to login? Even better.
User SegmentationThis new version now offers some user segmentation capabilities. For example, if you have users fill out a form, you can trigger on the data they enter, and break them into groups. Perhaps you ask people to indicate how long it is before they intent to make a decision. You might then want to track the behavior of users on your site who select 0 to 3 months differently than those who pick 3 to 6 months.
You can now do this through the use of code on your web site to set a variable for Google Analytics to key on. I have not yet implemented this to test it, but I hope to do so in the new future. But, it looks like a promising enhancement to the product's overall capabilities.
What's Still MissingThe Analytics Settings section of the product does not appear to have changed. As a result, the process for setting up filters is still the same. This is too bad, and hopefully we will see this change in an upcoming release. Filters can be used to setup "Content Groups", and I explain how to do this in detail in Google Analytics in my article Using Google Analytics to Manage Content Groups.
Content Groups are useful on larger sites where you want to look at the data across a larger set of pages. Let's say, for example, that you have a site that sells books, and you have organized them all into categories by genre. Let's also assume that all the genres have their own template that determine page layout because you have determined that different audience respond differently to the page structure.
If you have several thousand science fiction books, with one page per book, it is hard to look at the performance of any one page and draw meaningful conclusions about how your template performs in terms of driving conversion. There just is not enough data. This is a perfect application for a content group. With a content group, you can aggregate all of your science fiction pages into one page, allowing you to measure their performance in a simple and elegant way.
As I mentioned above, you can setup Content Groups in Google Analytics, it just requires you to define separate website profiles for each Content Group. The details are explained in my prior article about this. It works fine, but it just feels a little bit clumsy.