@SameerPatel It's the HP mini 311. Trying to get used to the smaller keyboard and screen. Very portable, easier to use on plane. in reply to SameerPatel 28 mins ago
«
»
Comments

Defining “Engagement”

Categories: Media 2.0, Social Media MeasurementPosted on February 1st, 2007

I’m into this whole defending terms and definitions crusade as I believe it defines where we go from here.

Don’t use that term until you know what they mean
That term Engagement sure appears to be thrown around recently, (and I’m not talking about Cap’n Jean-Luc Picard) with a wide variety of use and understanding. Honestly, it scares me quite a bit.

Some highly criticize this term as it’s being pushed by the advertising industry to move into this space. That’s bull crap, because we did a study ourselves (none if few were advertisers) at a a recent roundtable, and it scored as the top ‘wanted’ attributed to be measured, for whatever that’s worth.

A few weeks ago, I compared how Advertisers think of engagement vs a social media professional.

It’s important to define and measure
Before the snarks, haters, and other ill-tempered cube bullies jump on to attack such a silly notion, (I’ve recently fallen under attack for defending the term social media) please remember that in the end, we all realize that it’s impossible to truly measure something gray (humans come in so many different flavors), but our clients, bosses, and customers will hold us to at least try with whatever yard stick or tape measure we decree as patternable.

CEO of Nielsen Buzz Metrics Johnathan Carson discusses Engagement importance and measurement but does not define it.

Respected thought (and practice leader) Eric Peterson defines engagement as:

“Engagement is an estimate of the degree and depth of visitor interaction on the site against a clearly defined set of goals.”

He suggests that it will be measured depending on the needs of every client or situation. I noticed that pattern upon reviewing Charlene Li’s recent Blog ROI report she sent to me.

Whatever we decide for that term to mean, engagement is only part of the story to be measured, there’s quite a few other attributes that require study to accurately tell the story.

I find Eric’s activities (part two, and three) (Update Feb 12th: and now part four) taking on the actual measurement insightful and fascinating, this spurred from the challenge my colleague Robert Scoble requested of the industry to define and calculate engagement metrics. I’m kind of the middle of this thing, as I also have been communicating with Clint Ivy, and Daniela of Factiva, it’s a fun party I can assure you.

You can view all my thoughts that are tagged Social Media Measurement, and yes Brian S, excellent job!

Ok, I’m going to be brave and attempt again, I won’t be last, and I encourage anyone to make suggestions, here goes:

My working definition:

“Engagement indicates the level of authentic involvement, intensity, contribution, and ownership”

It’s possible for me to shorten it to:

“Apparent Interest”

I say apparent because someone can be interested and never act on it, measuring that will be difficult. If they act on it, say it, or gesture, then we can measure. I say interest, as I really see engagement the verb of interest.

Engagement Formula:

“Attention + Interaction + Velocity + Authority + Relevant Attributes (variable)”

This formula uses Attention, which is often time-spent. The attribute should be limited to when users time out of a website, as a single user can have multiple tabs open, and even open all night. Interaction is a clear example of a user being engaged. You need both attributes together as a user could be engaged without interacting with the site. I listed out “Other Attributes” as each tool is different with variable outcomes. For example user engagement on my blog could range from reading, commenting, or subscribing. Tools like Facebook could have a different set of attributes to determine engagement, such as poking, sharing, or friending.

Update: Feb 3rd
The Ad agency tried to define engagement over 8 months, they were criticized. Wiredset has some interesting concepts.

Based upon the great feedback and discussion in the comments, (Damon and Charlene Li in particular) I think it’s now possible to even simplify the term down to “interest”.

Update: Compete has now launched a service that is tracking web activity from a browser plugin (like Alexa) and has given a rating to the sites with the most ‘attention’. It will be interesting to compare this data to Alexa, MyBloglog, and Attention Trust type services.

Share This Post:
  • Digg
  • del.icio.us
  • Facebook
  • Google Bookmarks
  • Reddit
  • Technorati
  • LinkedIn
  • FriendFeed
  • StumbleUpon
  • Twitter

This entry was posted on Thursday, February 1st, 2007 at 3:51 pm and is filed under Media 2.0, Social Media Measurement. You can follow any responses to this entry through the RSS 2.0 feed. You can leave a response, or trackback from your own site.

Comments
  • Good discussion going on with a subject I am very involved with as well. However, my perspective is a little different in that what happens on the site can stay of the site - is that engagement? Rather, does the person 'leave' the content with an intention to act, or actually act differently - whether it be a purchase, social or health behavior. "Interest' bothers me a bit as I think interest is something people bring to a site that can then be doused, met or ignited. If they have no interest, they are probably not at the site to start with.

    The engagement dimension is much like figuring how to 'motivate' someone; in fact, people already have motivations, the question is how to tap into them. People have many different interests, at various levels or intensities. What engages this interest towards a mutually beneficial goal (for the content producer and the user)seems to me to be the question.

    Here is a summary of some of the work I am doing. Would appreciate some comments and feedback.
    http://socialmarketing.blogs.com/r_craiig_lefeb...
  • My understanding of engagement is experiencing an emotion and acting on it - expressing it, reflecting on it and sharing it with others.
  • Interactivity would be encompassed within the Physical and Emotional scores. On its own, the term 'interactivity' doesn't mean anything; it must be given context.

    The Sony Bravia bouncing ball ad required little physical activity to consume, but (for me) was a powerful emotional experience.

    Alternatively, the M&M online film quiz in October 2006, promoting its new dark chocolate product launch, was highly interactive but left me wholly emotionally disconnected from the brand.

    Those activities which combine the two are memorable: Subservient Chicken is but one example. Interactivity must be contextual; it alone isn't a measurable distinction.
  • This is good David, how are you measuring the interactivity in an instance of 'engagement'?

    I think that somehow belongs in the formula. I'm not quite ready to talk about the formulas yet, just trying to weed through all the definitions first.
  • Maybe we approach it the same way we rate a diamond, where each stone is assessed on key qualities: colour, clarity, cut, carat and cost.

    We can talk about what characteristics we should use for marketing. I think we need to develop a system which takes into account all the elements we've been discussing. I still believe any measurement *must* include a human metric whilst attempting to gauge depth and reach.

    I'm open to discussing these, here's my initial thoughts on engagement (scale):

    Reach (1=low > 10=high)
    Influence (1=light > 10=heavy)
    Physical (1=low > 10=high)
    Emotional (1=lower > 10=higher)
    Value (1=low > 10=high)

    Using this common scale, we can develop a weighted rating to compare a variety of marketing activity. Of course, there are so many touch points, it is difficult to use this scale as a practical person-by-person guide, but perhaps a representative sample could be more useful.

    Thoughts?
  • In addition, I think Your visitors engagement can also be affected by extrenal factors.

    Today I read an article and move out without any thoughts - I am not engaged at this moment.

    Tomorrow I read another article on same topic that references your article, I revisit your article and now with the new perception that I have got by reading the new article, triggers a thought process in me about your article - I might even post a comment now.

    So first time when I read your article I am not engaged But second time when I revisit the article I get engaged to it and this second visit and engagement is actually caused by my new perception of the topic that I gained by reading the new article.Here I am not only engaged to the new article I read but it also works as a catalyst for my engagement to this article.

    and am sure, this is just one.. there can be multitude of such external factors that affects the level of engagement for visitors.

    Now, I think these makes measuring all the more difficult and more interesting ;)
  • When Damon Billian says,
    "a lot of other people don’t feel that lurkers are actually part of the community, something I entirely disagree with"

    I totally agree with Damon. What if a lurker reads this post today and tomorrow when he reads some related article he posts a comment there and refernce this article by posting a link to this page ?

    He is definitely "engaged" to that article because he is posting a comment there and he remembers what he read here and posts a link back to this page as reference- after that can we afford to say that the lurker was not "engaged" to this article ? - but if we are looking at his "apparent" action on this page(he read the post and left) he don't appears to be engaged at all.

    I think Charlene was putting up a similar point.
  • David

    I'm glad you agree with my definitions, awesome.

    Like Charlene, you are right. Currently, I don't have any detailed plans to measure emotional engagement for two reasons: 1) It's elusive and very challenging 2) When it comes to Social Media, I'm really looking for the 'effects' of what happens on the network.

    I also suspect there's other services that have been around for decades that are already measuring emotional engagement ("brand perception" studies for starters). I'm focused on this 'new' media toolset.

    Fair enough?
  • The accountancy approach is valuable; positive ROI is one tangible, audited benchmark we can use and agree.

    Yet I still maintain we are interacting with *people*, as such the sums must reflect a human metric. In this, Jackie Huba and Ben McConnell define four archetypes in their book Citizen Marketers (filters, fanatics, facilitators and firecrackers) borne from successful social media activity. This feels like a better starting place in our quest for measuring engagement, although I admit administering the census could be tricky.

    As an aside, should we be considering two axis? Emotional and physical engagement?

    I consider this because I bought a bottle of water yesterday before getting on the train; I chose it because of convenience and, for the life of me, I wouldn't be able to name the brand now. How engaged was I? Physically - yes, through the act of paying for it. But I had zero emotional connection.

    I would agree this wouldn't meet your above definition of engagement.

    However, we can't have it both ways. We cannot use conversion (eg. sales) as a metric of engagement and then discount it because it wasn't "engaging enough".

    That's being convenient in our counting, For an increase in revenue might please our client, but it may not necessarily represent increased engagement, a fine distinction not accounted for on the balance sheet.
  • Interesting Dennis, kind of like 'interest opportunity cost' or 'elasticity'
  • Another angle to measuring engagement relates to the willingness of the person to resist alternative choices. This relates not only to the individual's goals (idle surfing vs. targeted research vs. looking to buy something) but to the availability of alternatives ("stay here since it's the only source" or "stay here because it's better than the alternatives").
  • Of course, you're both right Damon and Charlene.

    Individuals can be become engaged without physically expressing it.

    Perhaps the reason why we're placing less focus on'non-apparent' interest is because it would be extremely more difficult to measure.
  • Thanks for adding your voice to the engagement discussion -- I've been following the debate and have been doing some thinking about it as well.

    I think you're getting somewhere with your definition, as it allows engagement to be indicated at the most minute level -- simply indicating some sort of interest. From an offline perspective, a print ad "engages" someone when it catches her eye -- her interest -- with a flashy graphic. So I absolutely agree with Damon that interest can be very passive and doesn't need to exhibit the explicit engagement that Eric Peterson was measuring in his series.
  • bess
    Jeremiah,

    Keep posting. I actually find some technical folks working on measuring AJAX interaction and Web 2.0 media. I will look into it after I figure out the Browser situation in WWW.

    Opera CTO, our Browser speaker, is the inventor of CSS. He is actively working on CSS3 standards. CSS3 has a few properties that control media. This is still at a very elementary stage. May be W3C need to understand the increasing consumer demands on audio/video/media.
  • Hi Jeremiah,

    Can you see what people click on in your recommended reading section?
  • We're making progress here this is fun.

    Have you read up on 'gestures'? One can be involved, engaged, and do something without leaving a comment.

    In google reader, you can choose to share an item, which is an indicator of interest (engagement)

    For example my "link blog" see right nav, is an example of a gesture, which indicates my engagement, although I never 'clicked' on anything in the post, or left a comment.

    Sometimes, I subscribe to other people's link feeds (like Scoble's) and feeds that I did not subscribe to end up in my link feed.
  • Hi Jeremiah,

    I was referencing WOM in an offline sense as well (something that isn't trackable). I guess I am doing the slant from ecommerce because that's what most of my background is;-)

    "Either way, I always welcome your opinions, and love how you leave comments on my blog, you are so ENGAGED with me!"

    Well, we aren't engaged in that sense...haha! Just be sure to pop by my neck of the woods sometimes;-)

    "Apparent Interest: To me, Engagement isn’t about clicks. It’s about ‘interest’ level. How long did they stay, what did they read, what did they think, who did they tell, what did they say, did the leave a comment, etc."

    Interesting....didn't think about it from that perspective. I guess you could also do it from a number of readers vs. number of comments perspective as well (I have xx# of readers & xx% actually leave a comment).

    I found the report to be interesting because a lot of other people don't feel that lurkers are actually part of the community, something I entirely disagree with because the number of lurkers will always outweigh "active" participants & their value shouldn't be diminished just because they don't leave a comment (they are, after all, still digesting information & forming an opinion based off of the content).
  • Damon, I spent some time reading the PDF.

    Here's a summary right from the abstract:

    "When compared to a
    control group, community participants bid more often, won more auctions, paid higher
    final prices on average, spent more money overall for buying items, and were more
    motivated to make purchases if they didn’t do so before. Community enthusiasts also
    listed more items for sale and earned greater sales revenues. However, lurkers followed a
    different selling strategy: they listed fewer items but converted more of them into
    successful sales."
  • Oh and Damon, I've started to see that Community site a few times in different locations, so I thought it was worth subscribing to.
  • Damon

    I can't tell for sure, but it sounds like you're thinking of Social Media Measurement and Engagement from an eCommerce slant. If so, that's different than the viewpoint I'm seeing.

    It indeed is very possible to measure the WOM of your rabid fans. If one of my rabid fans blogged about me, then linked to me, I can track that. I can then do some crude measurements on that persons authority (technorati rank, alexa, their own engagement) to see how impactful that was.

    Apparent Interest: To me, Engagement isn't about clicks. It's about 'interest' level. How long did they stay, what did they read, what did they think, who did they tell, what did they say, did the leave a comment, etc.

    Wow that article is very interesting. It was hard to tell what the outcome was from the summary. Is it that lurkers were more effective or active participants?

    Either way, I always welcome your opinions, and love how you leave comments on my blog, you are so ENGAGED with me! (and my personal friend too) ;)
  • Hi Jeremiah,

    I think tanglibles can be hard to come by for certain aspects of engagement. For example: there would be no way to track WOM by your most rabid fans that participate on your site.

    Apparent interest: I guess this is a workable term if you were interested in tracking clicks that resulted in a "transaction" of some sort (20 people clicked on this, 2 people bought, etc.).

    I did find this interesting about eBay communities...and I think it is somewhat relevant to the discussion....

    http://www.futureofcommunities.com/2007/01/18/c...
blog comments powered by Disqus
site design by studionashvegas proudly powered by WordPress