This is a controversial subject indeed.
On the one hand we have Avinash warning us that engagement is not a metric it’s an excuse. The argument being that we shouldn’t use ill defined engagement metrics as a proxy for something solid like a sale.
Then we have both Theo & Avinash agreeing that using engagement for segmentation purposes rocks (As Avinash put it in a reply to my last reply about engagement! 😉
And in this post I see agreement largely with what Eric Peterson wrote when he published his Engagement formula (also very controversial). Theo basically says “that a customers degree of engagement is better calculated as a synthetic metric composed of several basic metrics, rather than as a one-metric solution e.g. measuring customer engagement by means of duration of visit only.” And that is exactly what Peterson said in his formula – he calculated a number of things that an individual had done as an overall score in order to determine his most engaged visitors.
My own view is that you can define metrics across a dimension in a similar way that the Web Analytics Association discussed in their definitions document.
The WAA definitions state that a dimension is;
Dimension – A general source of data that can be used to define various types of segments or counts and represents a fundamental dimension of visitor behavior or site dynamics. Some examples are event and referrer. They can be interpreted the same as counts above, but typically they must be further qualified or segmented to be of actual interest. Therefore these define a more general class of metrics and represent a dimension of data that can be associated with each individual visitor.
The dimensions I’ve defined are Reach metrics, Engage metrics, Activate metrics and Nurture metrics. Now whether the exact purpose of the WAA definition is designed for this is not the point though I feel it’s exactly what the term was designed for if Jason Burby had anything to do with it. However the methodology that works. When you segment your metrics into these four dimensions then the semantic arguments around “what engagement is” become irrelevant as the whole lifecycle is defined. Using this methodology which has evolved in 3 different places I’ve personally seen, all of the above arguments work.
REAN (reach, engage, activate and nurture) was developed in 2006 by Satama.
HP designed an incredibly similar model which they discussed in October 2007 at the eMetrics summit in Stockholm. Nokia discuss a similar model (though admittedly they have worked with Satama for a long time).
Jason Burby & Angie Brown then published the above dimension statement as part of the overall web analytics definitions.
REAN is defined as such;
R – Reach sources, defining the methods you use to drive traffic to your website or online presence and measuring how people find your brand, product or service.
E – Engage, is the click depth and time spent interacting with your online creative elements and processes.
A – Activate contains the verb ‘act’. It means the visitor has taken an action on your website, preferably one that you had pre-defined and wanted them to take.
N – Nurture is the way you actively encourage your activated visitors to come back and consume more of your website content.
By then developing KPIs and segments within these 4 dimensions you consider all arguments. CRM and customer loyalty, latency, recency, all drop into the Nurture metrics, conversion drops into the Activate metrics, click depth and duration including funnels and processes all fall into Engage metrics and Reach are simply the sources of traffic. The effectiveness of reach is measured against the other 3 dimensions as your goals and objectives dictate.
In Eric Petersons formula you can see every one of the elements he uses fits into one of my dimensions. His was the first physical manifestation of the whole customer lifecycle as measured with REAN that I’d seen published, which is why I said that it was the most comprehensive measure of engagement that I’d seen at the time though I’ve never had to use anything that complex with my clients to date because of the REAN model.