Audience engagement is a fundamental micro-conversion for any content marketing campaign: it indicates that people are visiting your website for a reason and not just because they have stumbled across it.

With social media how you measure engagement is set by each social medium, such as followers on twitter, likes on Facebook etc. These measures of interaction are black and white: someone has either viewed your post or not, they have shared it or not, liked it or not.

However, with a website you have scope to build a more nuanced picture of how engaged your audience is and the effectiveness of that engagement in generating business.

The purpose of this article is to explain some of the problems with measuring content engagement, and how, using some of the more advanced features of Google Analytics, you can significantly increase the insight available into visitor behaviour.

Measuring website engagement ­for the silent majority

Measuring discrete definable actions in response to website content is fairly straightforward. Event tracking or social interaction tracking can be used to measure public responses to the content such as the number of comments made, comments liked and social media shares initiated from the website.

But these public responses are the tip of the iceberg: for each visitor that responds publically there maybe hundreds or thousands that regularly return to consume the content and never respond in this way.

Assessing the level of engagement of this silent majority requires a different approach.

Because there is no direct measure of how engaged a person is, you need to use measures that act as proxies for engagement such as: time spent viewing the content, the number of website pages viewed and the frequency of visits. These represent the time and effort spent by the visitor on your website.

Time spent and the last page viewed problem

One of the big problems with measuring time spent viewing content, is that the content you want to measure the engagement with is often on the last page the visitor will view on the website for that visit. After all, if you have found what you came for, you won’t always want to look further.

Google Analytics as standard does not record the time spent on exit pages. Since it relies on the next page being opened on the same website in order to tell how long the page was being viewed for.

It may be that this last page is also the first page, in which case the visit is recorded as a bounce and you will have no idea if the visitor spent any significant time at all on the website.

This problem can be overcome by setting up a timeout event. The timeout function works by checking to see if a page is loaded after a specified number of seconds.

So, for example, if you estimate that your blogs take a minute to read, you could set it so that if the page is still loaded after 60 seconds an event is created and a time stamp recorded for the visit.

This would mean that if a page is the last one viewed Google Analytics will record a time of 60 seconds. It will also change the definition of a bounce visit, from a visit where just one page is viewed to a visit that consists of a single page view that lasts less than 60 seconds.

There is a lot of flexibility in how this can be set up: it could be set to recheck every X seconds, it could be set to check all or just some pages, it could use a different length of time on different pages.

The important point is not to record exactly how long a person visited but to establish that they were there long enough to read the content, even if they only viewed one page.

Frequency of visits

For each visit Google Analytics keeps count of how many times that visitor has visited (this is the count of sessions dimension) and how many days it has been since they visited (the days since last session dimension).

This information is very well presented in the Frequency & Recency report found in the Audience / Behavioursection of Google Analytics. It can also be applied to most other reports by using custom segments.

This provides an overview of visitor behaviour i.e. how often regular visitors return, how long they remain loyal for and the content that they look at. They are also useful for investigating the difference in conversion behaviour between frequent and infrequent visitors.

However, if you would like more flexibility and clarity in your analysis you can achieve this by updating a unique identifier for each visitor into a custom dimension. This provides the data needed to make detailed analysis of repeat visits either through custom reports, standard reports or by downloading visitor data into Excel.

The easiest and most readily available means of doing this for most websites is to load the unique identifier from the Google Analytics cookie into a custom dimension in your data.

Making the numbers meaningful

On their own, none of the proxy measures for engagement: time, frequency of visits or number of pageviews, is a reliable indicator of engagement.

For example, if a visitor spends a lot of time on a single page or a visit they might be finding your content fascinating or simply difficult to grasp, similarly if a visit consists of a large number of pageviews the visitor could either be very interested or unable to find what they want.

To form a proper understanding of your web visitors relationship with your website, you need to look at these measures in relation to each other and to take into consideration the nature of the content you are providing.

However, the most important step is to establish the relationship between the engagement metrics and other micro-conversions and, ultimately, the macro-conversions for the business.

game-changer-promo_2.jpg