The Art of Engagement
In my last post on Engagement Analytics in digital marketing I considered how customer engagement is typically measured, which is using Web analytics. A common way to calculate engagement is to compare the number of page views across two date ranges—before and after a campaign launch—and infer that the campaign is engaging if it produces more page views.
As I explained in the post, the problem with this method is that it conflates engagement and page views. Engagement is better represented by a concrete conversion action such as signing up for a newsletter or submitting a form, and moreover, it’s difficult to link increased page views to engagement (other factors, like unanticipated viral media attention, can create a spike in the statistic). The bottom line is that aggregate averages of data, like most popular landing pages or number of page views, are used to infer engagement and that inference isn’t always justifiable.
By tracking millions of data points in real time, tools like Sitecore’s Customer Experience Platform (CEP) help marketers build a more accurate picture of customer engagement—think Georges Seurat’s Parade de Cirque compared to Kazimir Malevich’s Red Square—and with relative ease. Sitecore's CEP is out-of-the-box, goal tracking is straightforward, data visualization of Business Intelligence (BI) analytics is elegant, and most importantly for our purposes today, there is a clear customer engagement model.
In Sitecore’s CEP, engagement is calculated by assigning a value to a goal using a differential point system. The goals are determined by your business objectives and the value assigned to each of those goals (on a scale of 1-100 in increments of 5) determines the value per visit.
Here’s a screenshot of a goal assignment:
The value per visit is the metric by which engagement (or “conversion potential”) is measured. As Jill Grozalsky at Sitecore puts it:
Viewing a report on value per visits helps marketers understand the importance and quality of traffic in relation to websites and marketing efforts being executed. If we are to compare two marketing efforts, say display ads and a social ad campaign, we might see that the display ad is producing more visits, but the social ad campaign has more value per visit—more conversion potential—which would allow us to make the decision to invest more in a social ad campaign because there is the potential for a higher ROI.
— Sitecore DMS Engagement Values: Attributing Value per Visit, Jill Grozalsky, Sitecore Blog, published 5/20/2014
Digital marketers can furthermore conduct A/B and multivariate tests on site components to determine engagement at a granular level, and personalize the site experience using this data as a means to optimize conversions.
The Art of Engagement
Sitecore’s Engagement Value model is an elegant solution to the problems of conflation and inference that plagues traditional digital marketing. It’s not free of complexity, though, and Sitecore understands this:
The marketing effort with the highest Engagement Value should be the goal that moves the visitor toward your strategic objective. If you pick the wrong set of goals to represent a strategic objective or you put an incorrect Engagement Value on the goals, then you will optimize your website in the wrong direction and you won’t reach your strategic objectives.
— Sitecore Engagement Value Best Practices, Ron Person, Sitecore White Paper
In other words, Engagement Analytics are only as useful as a marketer’s capacity to understand her business objectives, define the appropriate set of goals, and assign accurate values (or points) to those goals. This means that measuring engagement is still more of an art (or craft) than a science, which makes sense. I’m not of the mind that engagement can be defined purely programmatically, nor do I want it to be—see William Gibson’s The Peripheral for a convincing portrayal of the implications of machine learning.
Given the artful character of measuring engagement we are reminded that even elegant tools like Sitecore’s CEP are themselves optimized by our own skills and capacities. Marketers must be able to align their business objectives with analytical models, and the experiences that we design must themselves inspire engagement. In my final post in this series, I’ll consider how engagement is defined in another digital industry (video games) and how that model is relevant to designing engaging experiences for the Web.
About the Author
Leave A Reply