Monday, 21 November 2011

"Engagement": Accepting an Inadequate Metric

"Engagement".  It's a term that's often bandied about in media, and has been deconstructed and reinvented many times over.  However whilst within the digital sphere, conversation and debate over the concept of consumer "engagement" has evolved, consideration as to its appropriateness as a metric has somewhat fallen by the way-side.

Is it Engagement?  Or is it something else?

Recently speaking at the IAB Engage conference, Michelle Klein stated her belief that the currently used "engagement" metrics of dwell time, sharing, etc, don't accurately reflect the nature of a truly "engaging" experience.  Too often it is assumed that a consumer that plays a game, signs up for a Facebook competition or watches a video has had an "engaging" experience.  Realistically, the only certainty that can be taken from the resultant data is that the consumer interacted with the campaign.  Any meaningful brand connection as a result of that interaction is generally inferred.

Engagement goes beyond interaction or enjoyment of content.  It's an interaction that results in an emotional connection and association with the brand. 
Think of the last brilliant dramatic film you went to see.  One where all of a sudden you realise that your eyes hurt from watching the screen so intently.  Now that's engagement.  That's engagement to a point of personal immersion.  Whilst this isn't achievable (yet) at a campaign level, this example serves to show the extent of what a fully engaging experience can do to you.  This is what the goal should be.  Not the eye-watering part, but that feeling of unbiased personal connectivity to the previously unknown content put in front of you.  Judging whether a user achieved this connectivity through examining whether or not they entered their details or clicked through is clearly insufficient.  Generally, such interactions serve a more functional purpose, it taking a lot more to elevate them to the realms of becoming an engaging experience.

An Unfortunate Acceptance

It would be preferable to measure campaigns in terms of brand perception, consideration, preference, etc due to their superiority in measuring levels of a consumer’s engagement with a brand following content interaction.  Unfortunately however, the lack of science and direct accountability behind such methods renders them to be perceived as less efficient metrics than those currently used.

That said, despite the inappropriateness of current engagement metrics, they are far more preferable for branding campaigns than the more commonly used alternatives of CTRs and CPAs.

Why?  Because they fail to reflect the notion of what branding is.  Clients appear to be constantly asking for branding campaigns; though insist on measuring such campaigns on what are essentially performance metrics.  It'd be nice to say that this is solvable through a snazzy presentation on digital, but the reality is that it's not that simple.  Unfortunately, such demands are generally a result of those in the upper echelons of the client’s hierarchy stubbornly handing demands down to those that have no choice but to follow. 

Whilst a wish to focus on ROI is understandable, just as it would be fine to consciously state a wish to intermix branding and performance.  However to ask for performance metrics to be used on what’s been briefed as a solely branding campaign makes achieving communications goals tricky to say the least.  Not only does it make digital teams less likely to choose more "engaging" rich media over its more ROI-friendly brothers of standard display, but it causes a lack of creativity.  This generally results in a campaign that leaves clients scratching their heads as to why people still don't have a positive brand perception, or haven't increased their consideration levels.


So whilst the current use of engagement metrics is perhaps slightly unsatisfactory, if given budget to engage consumers in a digital landscape, I'd say "go for it".  Affording creatives and planner/buyers much more scope for creativity, it's a far superior branding metric to the currently favoured CPC/CPA system that's frustrating the industry.  Unfortunately, until either the client big-wigs get to grips of what a branding campaign is truly about (or those that do get their jobs), or engagement is measured in enough of a scientific way to sell it in to clients, engagement and branding will continue to be left in a confusing state.

1 comment:

  1. This is well done. I agree with the first half; however, not that current engagement metrics are preferable to CTR and CPA. Current engagement metrics are extraordinarily weak -- "likes" and "retweets" etc. have little connection with user intent, as you allude to in the top of this.

    The problem, as I see it, is we are trying to measure two paths:

    1. Direct response: Impression > Click > Sale
    2. Brand engagement: Impression > Interaction > Engagement > Rise in brand awareness or intent > Eventual sale

    The first doesn't work, because it neglects the huge additional value digital campaigns have in building brand engagement.

    The second doesn't work, because the steps from interaction to engagement (from retweet to really engaging) and from brand awareness to sale are both extraordinarily hard to parse.

    So we're stuck for now with direct response metrics that tell a portion of the story, and a brand engagement path that currently can't be measured.

    I think all smart marketers realize the overall pattern. And some of this can be parsed out via broader tools such as regression analysis, which look at overall lift from the entire digital prong as a variable against all other communication components.

    It's a thorny problem. If you can solve (2) above, you'd have a nice service, indeed.