Segments are a hugely powerful tool in a CRM. They let you create ‘dynamic’ groups (e.g. “People in Postal Code X”). Advanced segments opens up all of the contact fields, and more, as segment ‘conditions’. Creating an advanced segment that displays any pages that match your sub-directory name will show you any visits which included at least one page view on any one of the pages within that sub-directory. This definition – visits instead of pages from the previous paragraph – is an important differentiation. Advanced Segments are an incredibly useful feature of Google Analytics. They allow you to analyse subsets of your users, and compare and contrast behaviour. Google Analytics comes with a number of standard segments built in. (For example, New Visitors, Search Traffic, Direct Traffic.) However, the real power comes from leveraging your unique data to create.
Published September 1, 2016 | Categories: Google Analytics
I love Google Analytics. I love that you’re able to easily dig deep into data to discover hidden secrets about your website that might not be obvious with the basic reports in GA. In order to do this deeper analysis, you’ve got to use advanced segments. It can be scary at first for the average GA user, but once you start using them you discover how powerful they can be and the greater insights you can get from your data.
Recently, we performed a deep dive (aka some advanced reporting) for a client to try to uncover some of these hidden secrets about their website. We wanted to find out if certain features available to users was helping or hurting the sites performance. We could see that people were using these features, but needed to dig deeper to see how that usage affected goal completions and behavior.
Let me set the scene:
This client sells B2B (offline) and B2C (online). The goal of the website is to provide information about their products and also generate sales.
For our deep dive, we looked at many data points, but for this post I want to focus on only two:
- If the blog was helping to drive goal completions and qualified traffic
and
- How the retailer locator affected behavior on the site and goal completions
So to look at both of these, I had to set up advanced segments for visits including blog views and visits with no blog view.
Here’s a look at how those advanced segments looked:
Visits including Blog View
Visits with no blog view
Notice the only difference is one says “include” and the other “exclude”. Now let’s look at the data.
Look at the parts I highlighted. We see from a high level that visits with NO blog view have much better behavior. Sessions have lower bounce rate, more pages per session and higher session duration. Not only that, these visits’ goal conversion rate was 2.5x better.
So now you’re probably asking yourself, “What does this mean?” Well, it means the blog isn’t helping provide value for this site and increase the likelihood of a goal completion. No, they shouldn’t kill the blog and remove it; instead they need to look at their current blog strategy and how it can be tweaked and improved to help increase behavior and goal completions. For our analysis, we stopped at this point due to time constraints but when the client is ready to look at the blog and come up with a new strategy, we’ll dig deeper into this data and look at their top blog posts over the past year and how each of those has affected behavior. Doing that will allow us to get info on which blog posts provide value and which do not. Then we can come up with a strategy to refresh those older posts and create new posts.
Now let’s look at the retailer locator and see how it affects behavior and conversions.
Here is how I set up these advanced segments:
Sessions excluding Retailer Page
Sessions including Retailer Page
Looking at this chart, we can see the retailer locator improves behavior on the site and results in higher session duration, more pages viewed and lower bounce rate. However, it does NOT help ecommerce.
So what does this mean? Well it means, the retailer locator is a good feature to help improve session behavior but shouldn’t be as visible since it doesn’t increase online ecommerce conversions. Currently, the retailer locator was highly visible in the main navigation and the utility navigation. After looking at this data along with other data (not shown in this post) we determined it was best to leave this feature on the site but since selling online was the #1 goal of the website, we wanted to make it less visible and remove it from the navigations and move the link to the footer. This way the link is still accessible from all pages.
Something to note, with the retailer locator, we can only go off the data we have. Unfortunately, for this client it’s not possible for us to get offline information that we could then import into GA to get a clearer understanding of whether the retailer locator helps to increase offline sales or not and should be left highly visible.
As with any website change, it’s very important to annotate the change in Google Analytics so you can easily see if your change has caused any fluctuations in your data.
Now it’s your turn! Look at your website and see if there are things on it you would like to dig into deeper to see if it’s helping or hurting engagement. I bet you’ll be surprised and excited with the new data you’ll get!
And by the way, if you don’t feel comfortable digging into GA yourself, let Beacon help! :) We offer GA Training which can be customized to teach you how to set up Advanced segments and reporting or we can do the deep dive for you.
Recent
Posts
I think we can all agree that Google Analytics (GA) is totally rad. You get a wealth of website traffic data for the low, low cost of free. But sometimes the standard reports don't cut it, and in my opinion, if you're not using advanced segments, you're not using Google Analytics to it's full potential.
Advanced segments basically function as filters. They allow you to focus on specific types of traffic for deeper analysis. GA comes with a bunch of default segments built in, you can access them by clicking the Advanced Segments button on any of the standard reports.
As you can see the basic segments cover things like traffic source, traffic with conversions, mobile traffic and non-bounced traffic. All of these can be extremely useful, but the real power of advanced segments comes when you start to build your own.
Out of all the default segments, I've found Non-bounce Visits to the most useful. It cuts out a lot of junk and lets you see what those quality visitors are really up to.
To get started building your own segments, just click the New Custom Segment button in the bottom right corner of the Advanced Segments menu. This will present a new screen with options to include or exclude a huge list of metrics and dimensions.
Regular Expressions
Before we go any further, I'd like to cover a few basic functions of Regular Expressions (RegEx). In some cases, you would be able to build the segments I'm going to discuss without using RegEx, but it typically makes things quicker and cleaner. For example, if you wanted to make all social media traffic sources, you could do so by inputting 20 or so statements in your segment, or you could write one Regular Expression to match them all at once - more on this later.
Here are a few of the basic RegEx characters that you'll need to know:
- Pipe (|) - functions as 'or', e.g., Google|Bing matches Google or Bing.
- Period (.) - functions as a wildcard and can match any character, so 123.567 matches 1234567, 123a567, 123:567, etc.
- Backslash () - removes the RegEx function of the next character. You'll most commonly see this used in conjunction with a period for matching URLs and IP addresses, so facebook.com would match facebook.com but not facebook-com.
- Carrot (^) - indicates the beginning of a matching string.
- Dollar Sign ($) - indicates the end of a matching string, e.g., example.com would match example.com, help.example.com and example.com/blog, but ^example.com$ would only match example.com
RegEx can be frustratingly complicated, so I'll stop there. If you're interested in learning more, you can dive into some more detailed functions in this RegEx guide for SEO. The last thing I'll mention is that Regular Expressions are greedy. They are going to match as much as they can unless you tell them otherwise, so keep a tight leash on what you're trying to match.
On To The Good Stuff
Without any further delay, here are the nine Advanced Segments you can't live without.
Branded Keywords
We're starting off pretty easy here. To see only your branded keywords, select Include, Keyword, Matching RegExp and then enter your company name, any common misspellings of your company name or any other branded phrases that might be used to find your site in a search engine. Make sure you separate each word or phrase with a pipe. In this example, I used pronto|prontomarketing|monsoon.
Non-branded Keywords
Things get a little more complicated here. You'll need to include organic traffic and then exclude the same branded phrases from the segment we created above and exclude (not provided) them since these visits contain branded and non-branded keywords. Make sure you have an AND statement between including organic and excluding keywords. Unblocked games yeti hacked arcadeprehacks.
Only (not provided)
Having trouble figuring out what your (not provided) keywords might be? Click the link above to automatically add this segment to your GA account. Focusing only on your (not provided) keywords in the Landing Pages report can help you determine how many of these were branded searches and how many were non-branded.
Keyword Length
Sometimes there are hidden opportunities in your keywords that you don't notice when you're looking at a list of hundreds of phrases. Is there a long-tail phrase that converts better than your head terms? Narrowing down your keywords by the number of words in them can help you find these gems. Click the links below to add these segments to your account.
Note: Here's the RegEx for matching 3 Word Keywords: ^s*[^s]+(s+[^s]+){2}s*$ The most important part to focus on is the number in the curly brackets { }. This number should be one less than the number of words you want to match. So if you wanted to match 7 Word Keywords, the RegEx would be ^s*[^s]+(s+[^s]+){6}s*$
Page Depth
Page Depth counts the number of pages viewed in a visit, and it's a very good indication of visit quality. Building this segment is going to take two steps. First, we need to build a custom report that will allow us to compare Goal Completions with Page Depth. Enter some of your primary goals as metrics and select Page Depth as your dimension.
Take a look through the report and try to determine at which Page Depth most of your conversions happen. In my case, 80% of goal completions occurred with a Page Depth of 4 or higher. So those who viewed 4 or more pages during their visit were much more likely to convert than those who viewed fewer than 4 pages.
We can now take this information to build our segment. Essentially, we want to focus on those high quality visits with 4 or more page views so we can see where they are coming from, what pages they enter on and how they behave once they are on the site. To create this segment, you'll need to enter this information:
That Regular Expression matches any number from 4 to 99. If your research from the custom report shows that most of your conversions occur at a different Page Depth, say six or higher, you would edit the RegEx to read as follows: ^([6-9])|([1-9][d])$ This would then match any number from 6 to 99.
Social Media Traffic
Google Analytics has a full-blown social media section now, but I occasionally find that this still isn't detailed enough for me. So a quick segment can filter down to only social media sources and allow you to manipulate the data to your heart's desire. Click the link above to add it to your GA account. Here's the full RegEx that I use:
facebook.com|twitter.com|linkedin|del.icio.us|delicious.com|technorati|digg.com| hootsuite|stumbleupon|myspace|bit.ly|tr.im|tinyurl|ow.ly|reddit|plus.google.com|youtube|flickr|pinterest.com|^t.co$|tweetdeck
Take a look through your traffic sources and make sure I haven't missed any social networks that are sending visitors to your site.
Go Experiment!
Now get out there and have some fun! Experiment with using multiple segments at once. Non-branded Keywords with Page Depth sounds like it might yield some interesting results. Once you start feeling comfortable, try making your own segments. Got a video on your site? Set up Event Tracking and then create a segment that only includes visits with a video plays.
The possibilities are endless!
Advanced Segment Dac
If you liked this post, you might also enjoy Use Google Analytics to Develop Social Media Strategy