As has been a common theme throughout my posts on these GA features, segmenting your visitors is vital. Looking at your data all together tells you averages about how your visitors behave, but some of those visitors are valuable and some are not. What’s the difference? How do we make the route to a sale easier for the valuable customers? What is turning away the non-valuable ones?
Last time we saw how to segment your visitors based on whether they had visited a key page. The example we used was an add-to-cart page.
But what if your add-to-cart button doesn’t take them to a page, but instead just loads the item into a box on the side of the page? What if on your contact page are mailto: links that literally just open the user’s mail program, rather than launch a form? There should be a way to track these.
Well there is.
With event tracking you can tell Google Analytics that a user has interacted with your site in a way that doesn’t load a page. There will be no page loaded, so no _trackPageview() function call, so no data passed to Google. Unless you insert a new function call on the button/link/image/etc that you want to track.
_trackEvent(category, action, optional_label, optional_value)
_trackEvent() allows you to tell GA that something has happened, but without using trackPageview() to make Google believe a page has been loaded. So you can call this function and GA will provide information about these calls in the Event Tracking section of the Content report. You can add any labelling you like to the category, action and label variables.
Importantly, you can then use this information in an advanced segment.
For example, the screenshot above shows the setup for a segment where a user has downloaded a pdf, but not the one labelled “general”. This might be used to track people who request brochures for specific holidays for instance. The user could have downloaded one of any number of brochures, but we know they didn’t download the general company information pack. This is a useful segment, because our PPC traffic could be going straight to a holiday specific landing page and if the person downloaded the pdf with no confirmation page then they could look like a bounce when they have actually interacted with the site.
Now, in a similar way to with general key page view segments, I can compare visitors who did download against visitors who didn’t, to determine if there may be any technical problems for some, rendering issues in some browsers, or if I’m sending too much of the wrong kind of PPC traffic. The usefulness of this is far beyond what I can see by viewing my visitors in aggregate, and I now have the ability to segment these visitors by views of key pages or by other types of interactions.
For more information on this new Google Analytics feature, watch the official Google video.
In the next article, tracking site searches that return zero results.
Share this article
The latest posts delivered to your news reader...
Pulley!
