Google Analytics - A starter guide
Understanding how audiences are engaging with your narratives is vital, not only to assess where their interests lie, but also to inform the design and content of your future stories.
Below are some of the basics that may be of use when digging into the data collected in Google Analytics for a story created in Shorthand (so long as you have a Shorthand plan that supports Google Analytics and have either inserted your GA tracking ID under Settings, or had your tracking ID added to all your stories via your custom theme).
When you are digging into the data, ensure that you are searching the correct date range, which you can change in the top-right corner of the GA interface.
Understanding session vs page metrics
When getting started with Google Analytics, it's important to understand the difference between session and page metrics.
A session is a group of interactions that a single user has with your website. During a session, the user may visit a number of different pages on your site, triggering multiple page hits - called Pageviews - in Google Analytics. Session metrics calculated by GA include Avg. Session Duration and Pages/Session.
Page metrics relate to interactions of the user with a single page on your site. Some examples of page metrics are Avg. Time on Page and Unique Pageviews.
A particular strength of Google Analytics is understanding user behaviour over a session. However, because of the way Google collects information on single pages, Google Analytics has some limitations when it comes to tracking page metrics - particularly for single-page sessions, or for the last pageview in a session. It is outside our scope to cover this in full here; however, if you are interested, you can read more about how Google logs Sessions and Pageviews.
In your GA account, page metrics are predominantly found under the Behavior tab in the left sidebar. Statistics such as Pageviews and Avg. Time on Page can be found under Site Content > All Pages, and page load speed is located under Site Speed > Page Timings.
Avg. Time on Page
Dwell time is one of the key metrics in longform storytelling, and it is often said to be a better reflection of true engagement than Pageviews. Unfortunately, Avg. Time on Page is one of the least reliable metrics in Google Analytics, as a result of the aforementioned challenges that Google has in tracking single page metrics.
Locating Avg. Time on Page for your story
Under reports, navigate to Behavior > Site Content > All Pages. You'll then need to locate your story, and the easiest way to do this is by searching on Page Title - therefore, you should select Page Title as the primary dimension.
Type your title into the search box and then hit enter.
One benefit of organising the information by the Page Title dimension, rather than the Page dimension, is that it consolidates all of your story data regardless of the presence or absence of trailing slashes, or of tracking parameters that appear after the page path, such as UTM parameters and Google or Facebook parameters. Otherwise, a single page can appear in many different URL permutations, as shown below. (Of course, if you have two or more pages with the same title, or your story has had multiple titles over its lifetime, you will need a different strategy!)
Avg. Time on Page is intended to provide the average (i.e., mean) duration of all Pageviews. You should bear in mind that Time on Page has no way of excluding cases in which the reader opens a story and then leaves it open in a tab while engaging in a different task, later returning to the open tab. In that sense, it does not necessarily equate to active engagement. In addition, there are some challenging technical issues with how Google calculates this metric, as outlined below.
Cautions for Avg. Time on Page
Avg. Time on Page is computed via a legacy formula that assumes that no timing data is available for any exit visit (i.e., any visit in which the session ends at the current page - either because the visitor closes the browser or tab, or moves on to another website). This assumption was valid in the early days of the web, but it does not hold any longer in the presence of in-page interaction events.
All Shorthand stories by default trigger GA events, including video, link and section linger events. These events provide our storytellers with vital information about reader engagement with story elements. Though these interaction events are configured exactly as recommended by Google, the very fact that they result in logged Time On Page for exit visits means that Google's own Avg. Time on Page metric becomes inflated. This is particularly the case when the Exit Rate (% Exit) is high.
If Avg. Time on Page is an important metric for you, we recommend that you configure a custom formula to calculate a more accurate average by dividing total Time on Page by non-bounce visits instead of non-exit visits. For instructions on how to do this, we recommend this informative article from PACE on a new approach to Average Time on Page.
Alternatively, you can redefine and analyse the Avg. Time on Page metric within an external analytics tool hooked up to your Google Analytics account, such as Google Data Studio. We have a Google Data Studio report currently under development, which will not only incorporate the non-bounce Avg. Time on Page metric, but also a fresh update to our Google Analytics Section Linger report. If you would like an advance copy of this report, please get in touch.
Finally, we would like to add a word of warning about comparing Avg. Time on Page across pages that have different event configurations. It is reasonable to compare this metric across Shorthand stories, but you should exercise caution when comparing the Avg. Time on Page of a Shorthand story with that of a page that contains no interaction events, or a different configuration of interaction events. It has the potential to be an apples and oranges comparison, given the complex ways in which interaction events can directly modify basic metrics like (total) Time on Page and Bounce Rate - and, indirectly, derived metrics such as Avg. Time on Page.
Shorthand stories are configured to log three types of GA event that can provide information about how visitors are interacting with your content:
- section linger events (Event Action: linger-seconds)
- video events (Event Action: play / pause / end)
- outbound link events (Event Action: click)
Section linger events
Section linger events allow you to see where people are spending the most time in your story. A section linger event is fired each time a section is in the viewport for at least 5 seconds. For detailed information about how to locate and interpret these events, please read our support page on the custom section linger report for Google Analytics.
Video events relate to native videos with explicit controls (not, for example, background videos in your Title or Text over Media sections). To see how many times visitors have interacted with your videos, go to Behavior > Events > Overview and click on the Event Category "Videos". To restrict the data to the Shorthand story of interest, add Page Title as a secondary dimension, and then use the advanced settings link at the top right to restrict Page Title to the title of your story.
Set the filter to include pages containing your title, and then hit the Apply button.
Next change to the Event Action view by selecting that tab.
This will allow you to see the breakdown of play, pause and end events as pictured.
Finally, if you have more than one video in your story, add Event Label as a secondary dimension to see the events grouped by video.
If you require video engagement statistics often, you may wish to define a custom report in your GA account to simplify the process of organising video event metrics by story - or, alternatively, contact us for an advance copy of our forthcoming Google Data Studio engagement report.
Outbound link events can show you how visitors are interacting with the links in your story. Go to Behavior > Events > Overview and click on the link Event Category. To restrict the data to the Shorthand story of interest, add Page Title as a secondary dimension, and then use the advanced settings link at the top right to restrict Page Title to the title of your story - exactly as described earlier for video events. Finally, change to the Event Label view to see the breakdown of link clicks by URL.
To better understand the GA events your visitors are triggering as they move around your site, you might like to spend some time looking at live analytics - accessible via the the Real-Time tab under Reports.
Open your Shorthand story in another browser tab and then watch your page hits register in the Real-Time report. To see the linger, video and link events flow into the view, navigate to Real-Time > Events. (This assumes that you aren't filtering out your own network traffic/IP addresses in the View settings).
You should bear in mind that there can be a 24-48 hour delay before page and session data propagates to the non real-time reports.
Traffic sources & referrals
To find out how visitors have arrived at your Shorthand story, locate your page under Behavior > Site Content > All Pages, using Page Title as the primary dimension, as described above for Avg. Time on Page.
To find the Source and Medium (e.g., google / organic - representing organic search traffic from Google), search for Source / Medium as a secondary dimension.
To find the full URL of referring pages, set the secondary dimension to Full Referrer. The results will then show, in order of highest to lowest, the referrers of traffic to your story.
To dig into demographic data, in the same view as above, return to the secondary dimensions list, and choose from the options listed under Users.
These options include the browsers used to view your story, as well as information about readers' geographies (city, country, region or continent). This information can be helpful when deciding how to target story testing for available browsers, and how to time the promotion of your stories.
Under the user dimensions you can also see which devices readers are using to view your stories. Selecting Device Category as the secondary dimension, you will see a breakdown by desktop, tablet and mobile. Selecting Mobile Device Info, you will see specific brands and models of phones and tablets used by your readers.
Again, this information is vital for ensuring you are building experiences designed for how your audiences are reading.
Navigation within multi-story projects
If you have created a multi-story project in which all pages are configured to contain the same GA tracking ID (highly recommended), you can also track movement between pages.
When viewing a page’s data within Behavior > Site Content > All Pages, choose the Navigation Summary tab to see which pages with the same tracking ID were visited before and after the one in view.
The data will tell you how many views originated from a page with a different or no tracking ID or via direct hits (Entrances), and how many from pages with the same tracking ID (Previous Pages). Underneath you can see how many Pageviews resulted from each referring page.
You can also see how many views led to a subsequent page with a different or no tracking ID, or no subsequent page (Exits), and how many moved to another page with the same tracking ID (Next Pages). You are also presented with a list of each of those pages, with the number and percentage of Pageviews.
If you don’t have a multi-story project, but are trying to direct readers to other stories on your site at the end of your narratives, this can also be a helpful way to measure the success of that strategy - provided you use the same tracking ID across those stories. If you aren't using the same tracking ID, you can use outbound link events as described earlier.