> I've been able to simplify it down to how it matters to web folks:

> Engagement is "Apparent Interest"

> While it's been argued that one can be engaged and never measured, I
don't care, as we won't be able to measure! It's "Apparent" because
they've done something on the site, or paid attention, or interacted
with the site, or even a casual gesture.

Right. It's really not worth measuring something if you can't take any
action of it, and to be worth taking action on, it has to have some kind of
measurable value. Otherwise, there is no point in measuring it.

But, "done something" or "paid attention" are actions without a *time*
element, and I don't think you have "engagement" on an action alone. If so,
you are saying there is no difference between someone who paid attention for
30 minutes and someone who paid attention for 30 seconds; no difference
between someone who posted last month and someone who has not posted in a
year. Is that really true? I don't think so.

Time can be used to describe the duration of an action, and also can be used
to describe how long it has been since an action. Both describe how
involved someone is with the action, how engaged they are.

Engagement has to capture this idea of "being involved", or it's not really
any different from all the other stuff people currently measure. It has to
capture the idea that someone who is involved is more valuable than someone
who is not involved. And the reason is this: someone who is more involved
is more likely to create value for the company *in the future* than someone
who is less involved.

The value of having this knowledge is not very hard to imagine, at least it
shouldnt be hard to imagine for marketing people.

Nielsen / comScore didn't make a change like this for the heck of it, there
is something driving it, and that means the agency / media buying complex.
The money, in other words. And that money, collectively, is probably smart.
In this case, I think there is pressure to get on with applying the same
standards to web media as are applied to other electronic media.

At this point, "duration" is probably the best these folks can do with the
time variable, and for the advertising business, it's not hard to imagine
longer duration = greater potential for ad effectiveness. Whether that is
true or not remains to be seen, but it makes sense, especially since "time"
is used in the measurement of other electronic media - it's a standard.
That doesn't mean I'm endorsing "duration" as an engagement metric for your
business, it just means duration is appropriate for the display ad model.

If someone who is involved is more valuable than someone who is not
involved, there are a lot of good reasons to measure engagement, the primary
one being Management cares more about the future than they do about the
past, because they can "do something" about the future. As an analyst, if
you start reporting on the future - predicting - as opposed to reporting on
the past, people are really going to pay attention.

Predicting the value of Content, Widgets, Products, Campaigns - that is a
capability Management will be quite interested in, I think. Because they
can "do something" based on those predictions.

So, for example, using the other time metric, "time since", you can measure
these areas for engagement value:

1. Content Areas
2. Widgets
3. Products
4. Campaigns

Which content areas, widgets, products, campaigns create visitors /
customers who remain engaged? Which do not?

Then within each area, rank the different assets according to which ones
produce the most engagement. And the reason you want to do this? Because
engagement represents the creation of future value to the company, and this
value can be measured over time just like the value of any other asset. You
want to increase investment in areas that create involvement, and decrease
investment in areas that do not.

The content producing the highest engagement should get the highest
investment, the campaign producing highest engagement should get highest
investment, and so forth. It's prediction of ROI, as opposed to backward
measurement of ROI. And as the future plays out, you can measure the
accuracy of your prediction and optimize. It's really no different than all
the other "optimization" work people analyze every day.

Here is an example using social media. Say you have a bookmarking site and
the number of accounts and total bookmarks and so forth is rising (volume),
but the percentage of dormant accounts (say, no activity for 2 months) is
rising more rapidly. At some point, the dormancy issue overpowers the
volume issue and the site starts to spiral downward in volume. You can
predict this downward spiral will happen by looking at the "time since"
metric and try to take action before it's too late.

Most people look at volume as a measure of popularity and growth and by the
time volume starts dropping, it is already too late to save the site. The
audience *has already* dis-engaged. The engagement drops off before the
traffic volume does, and that is why enagagement is predictive and directly
addresses future value.

I'm sure you can all think of various start-up sites where this is happening
right now based on what friends are saying and general buzz on the boards,
and you think to yourself "they're not going to be around in 2 years".

You made a prediction. You didn't need any fancy metrics, just a sense of
engagement. If you asked these friends "how long it has been" since they
interacted with these sites, they would probably say "a while".

If we want engagement to be a measure that has value and is actionable,
there has to be a time element. Otherwise, it's just counting events that
take place, and we already collect a ton of that information.

If we want to create something valuable out of engagement, it can't be just
a more complex version of what we already have - visitor did this, did
that - it has to address a new dimension of the behavior - how engaged are
they in the action, how likely are they to do it again?

Jim Novo
jim@...
Web Site: http://www.jimnovo.com
Blog: http://blog.jimnovo.com/



--- In webanalytics@yahoogroups.com, Jim Sterne <jsterne@...> wrote:
>
> At 08:09 AM 7/11/2007, jchasin@... wrote:
> >-- I'm wondering what folks here think engagement actually is/means?
> >How do you define the concept of
> >engagement...
>
> While I wouldn't go all the way to "poppycock" I do
> spout off about Engagement in this podcast:
> http://tinyurl.com/yv7zwy
>
> >The ARF definition of engagement is, "Turning on a prospect to a brand
> >idea." I haven't bought into that one quite yet.
>
> ANA, AAAA, ARF
> Engagement is turning on a prospect to a
> brand idea enhanced by the surrounding context
>
> Jeremiah Owyang
> Engagement indicates the level of authentic involvement,
> intensity, contribution, and ownership
>
> Eric Peterson
> Engagement is an estimate of the degree and depth
> of visitor interaction on the site against a clearly
> defined set of goals
>
>
> ------------------------------------------------------
> eMetrics Marketing Optimization Summit
> Washington D.C., Oct 14-17
> http://www.emetrics.org
> San Francisco - London - Dusseldorf - Stockholm
>
> WAA Base Camp
> http://www.emetrics.org/waabasecamp/
> July - August 2007
> Los Angeles - Boston - Chicago - New York - Washington DC
>
> Gurus of Marketing Optimization
> http://www.emetrics.org/gomo/
> July - August 2007
> Los Angeles - Boston - Chicago - New York - Washington DC
> ------------------------------------------------------------
> Jim Sterne <jsterne@...>
> http://www.targeting.com +1-805-965-3184
> Chairman, http://www.WebAnalyticsAssociation.org
>
>
> [Non-text portions of this message have been removed]
>