Rebecca Cox - UX Designer
Menu button

Slicing and dicing with Google Analytics


GA has a heap of easy to use tools to filter, group and segment the usage data for a website or app. Its really important to use these to look more deeply - by relying on the reports as they're presented by default you treat your audience as if they were all the same, and miss opportunities to identify actionable trends.

Slicing and dicing your data can help you to identify behaviour patterns for specific groups of users, and understand the usage of each type/topic of content within a complex site. This enables resources (for example advertising, design and content development efforts and budgets) to be targeted where they will deliver the most benefit.

Use Segments to filter right across most reports

By default in Google Analytics, you'll be looking at data for all traffic to your website or app. (Unless there are filters set up on the view*)

For any indepth analysis you will find it useful to look at just the data for a section of your audience or traffic, for example:

Also, if there have been any unusual spikes in traffic (eg a link that went viral) it might be a good idea to exclude this from a general analysis, or look at it separately from other traffic.

Adding Segments

If you click on Add Segment at the top of the reports interface, you'll open a tool where you can apply premade segments or create your own. Segments can be created for a huge range of visitor behaviours, attributes and actions, from demographics, to pages visited, traffic sources, and even sequences of user interactions.

Once active, the Segment will be applied over most reports as you step through the Google Analytics reporting screens - so for your segment, you'll be able to see how they arrived at the site, what technology they were using, what pages they viewed, and much more. Be aware that a few reports won't have the segments applied (for example In-Page Analytics in full-screen) - so check before using the data.

Google Analytics Help pages - More information about Segments

One thing that I'd love to be able to segment on easily is the organic search query data that you get from a linked Webmaster Tools account (under Acquisition > Search Engine Optimisation). This would really help with digging into behaviour for different types of organic traffic, but isn't possible using the standard reporting interface. An alternative is to segment on organic keywords, however this doesn't include all of the "keyword not provided" organic traffic which the Webmaster Tools data gives access to.

* Where there are Include or Exclude filters set up on the View you are looking at, the reports will by default already have some traffic included or excluded. This type of filter is generally used to make the View more useful, for example to exclude staff traffic from the reports.

Get beyond the "top 10" by using the filter tool on reports

For many individual reports (including anything from browser names, search keywords, pages viewed, events) you will be presented with a table of data which may have thousands of rows, each with very small numbers of hits, with the top 10 rows displayed by default.

In many situations, this list won't by itself be saying anything terribly useful. For example, if you are looking at the top 10 Landing Pages for your website and these only account for a small percentage of all the website traffic, you'll need to do some work to extract meaningful information about the most common landing points into your site.

Try to identify ways to group the website content by URL, for example if the website has a large number of sections which all follow the same URL structure. Alternatively you may be able to group pages by topic keywords in the URL.

To apply a filter, click into the search box at the top of the data table and enter your keyword or query. You can also use the advanced tool to help set up more complex queries.

Using this technique you can group the data, for example you might discover that:

You'll be able to set these up much more easily if you understand even a tiny bit of regular expression syntax - Find out how to use regular expressions to set up filters - Google Analytics Help

Using Google Analytics Content Groups

Content Grouping is a feature in Google Analytics where you can identify groups of content on your site, enabling a similar type of aggregation and analysis as outlined above. Content Grouping must be set up for the site (in either / both the tracking code and Google Analytics admin) and is something that you would set up for ongoing use, rather than one-time analysis for a specific purpose. More about Content Grouping - Google Analytics Help

Using Regular Expressions to understand visitors screen resolution distribution

One report which can be hard to read is the one for Screen Resolution, as there are often hundreds of similar values given in the reports.

To break this down, first try using a filter on the report to see what the range of screen resolutions is:

Is anyone using a screen that is at least 3000px wide: ^3[0-9][0-9][0-9]

At least 1800px wide, up to 2999px wide: ^1(8|9)[0-9][0-9]|^2[0-9][0-9][0-9]

Then, once you have identified useful ranges to look at, set up a Segment using your regular expression to isolate each group.