• Kevin Cobourn

    The thing with engagement is … often times it is not measuring anything beyond the intend of a visitor. For example, let’s say we find people who engage with video convert at a higher rate. While that’s a correlation, it’s likely that the person who’s interested in video and engages is already more like to convert. In other words, engagement doesn’t predict revenue so even in correlations this metric fails.

    As an alternative, you can A/B test a video being on or off a page. Then, any change in conversion you see about that specific product could be more defensible in terms of proving causation.

    However, even THEN you can’t extrapolate that assumption to all videos … you see, it does get complex huh?

  • http://conversionscientist.com Brian Massey

    Kevin, it does get complex. This doesn’t mean that we can work with simple conclusions from what we learn. It’s smart to bet on statistical swings in conversion rate. The alternative is guessing or gut feel. Which do you want to hang your future on?

  • http://www.hennessysview.com Bill Hennessy

    I love this article. Even though the word “engagement” appears in my official title, I see, every day, how blind pursuit of engagement can undermine a company’s purpose. Thanks for debunking the myth and replacing it with useful information.

  • http://www.facebook.com/Jamesschramko James Schramko

    I have has similar results in my tests. Conversions are higher even with lower engagement.

  • Filip Galetic

    KISS – Keep it simple stupid. This mantra works every time. Good to see you delve more into how videos can drive engagement, I think this is a good story which illustrates some of the pointers: http://blog.treepodia.com/2012/06/toolking-com-boosts-conversions-up-to-60-using-treepodia-ecommerce-video-platform/

  • Carol Dodsley

    Great article and so glad to see someone split testing and measuring engagement V conversions within videos too – thanks for some great info

  • http://conversionscientist.com Brian Massey

    James, it would be great to see some of your data.

  • http://conversionscientist.com Brian Massey

    Thanks, Carol.

  • Andreas

    My experience is that the tool Visual Website Optimizer is measuring Engagement not very precisely. At least it didn’t a year ago.

    It does not count
    – form submits
    – actions on elements added after page load (e.g. click on an AJAX search)

    In my experiments customers often used the search more often, but it isn’t engagement in Visual Website Optimizer. You will say: OK, but the user will generate a page view after the search (e.g.the product detail page) and so engagement will go up, but VWO doesn’t count this, engagement does only count the actions on the tested page.

    In our tests with lower engagement but higher conversions our customers used the search more often and found their products faster – and so converted more.

    Tip: Use Custom Goal to track AJAX searches and / or define the Search Result Page as a goal.

  • http://conversionscientist.com Brian Massey

    Andreas, there are lies, damn lies and analytics. I appreciate you warning us about the VWO engagement metric. It’s something we’ve watched with curiosity, but have never paid much attention to. Thanks for the tip on search results pages. SERPs are SO important.

    Thanks for the comment.

  • Andreas

    Hi Brian,

    I meant the internal search of a page, not Google SERPs.

    Best regards

  • http://conversionscientist.com Brian Massey

    As did I. iSERPs?