A Frank Conversation about Social Media Measurement – Director’s Cut

This is today's Social Media Insider column, a complete version several hundred words longer than the abridged edition, which you can find at MediaPost

Bright lights, big podium - view from #ommasocial stage on Twitpic
The view from the OMMA Social stage

A Frank Conversation about Social Media Measurement

Everything you need to know about social media
measurement was covered in 45 minutes at MediaPost’s OMMA Social event in San
Francisco last week.

Okay, that’s not entirely true, but the five panelists I
had the honor of moderating covered a lot of ground and they probably could
tell you everything that matters – if you gave them just a little more time to
do so. There is a lot of ground to
cover, as you might have inferred from my column on 100 ways to measure social media.

Instead of trying to rehash the session, I’m letting the
panelists share their thoughts in their own words. Below are a series of
questions that I asked during the session, I planned on asking, or the audience
asked. Note that I shortened this version somewhat for readability; the
‘director’s cut’ is on my blog.

The panelists included:

  • Carla
    Bourque
    VP of Sales and Business Development,
    Six Apart
  • Steve
    Kerho
    SVP, Analytics, Media & Marketing
    Optimization
    , Organic, Inc.
  • Amber
    Naslund
    Director of Community, Radian6
  • Bill StephensonVP,
    Social Media Sales & Client Service
    , Nielsen Online
  • Jim
    Sterne
    President, Target Marketing of
    Santa Barbara


Not everyone was able to get their answers in before the deadline, but Carla,
Bill, and Jim have all shared their candid thoughts below.

When a marketer wants to know what
the value is of a Facebook fan, a Twitter follower, or a blog commenter, what
do you tell them? 

Jim:  It’s much
more than a person on the freeway not looking at your billboard or the person
in the bathroom while your TV ad blathers into an empty room. On the other
hand, once you link reach and awareness to conversion, you can do the math and
tell me.

Bill:
Placing a specific value on a fan or follower or commenter is a slippery
slope.  At Nielsen we believe the best currency to use when valuing fans,
followers, commenters, etc, is advocacy.  This is a metric that can be
calculated across most social networking sites and one that most large
companies can relate to and value based on their unique industry, product,
transaction price, etc.   A good social media partner can help their
client develop an actionable advocacy metric.

Carla: The
"value" of a fan is directly related to the relationship that that
fan has with the brand, and with the broader audience of influencers.
 Fans may have a certain value today based on their level of engagement
(their actions via passed links, demo and psycho data, etc.) but they can
be even more valuable if you keep them engaged and leverage them as a source
for advocacy.  Engagement, Amplification, and Equity are the three
highly-connected measures for determining value. The more relevant content you
can give fans to respond to and interact with, the more likely they are to get
involved, and their involvement will begin to ripple throughout their network
or social graph. At Six Apart, our social publishing heritage has taught us
that, the more marketers think like publishers – that is, continually offering
their fans content that keeps them engaged in the conversation – the more their
audience will respond and share their experiences. Ultimately, it is the
quality of the followers and fans, not the quantity, that is most valuable.
Interestingly, you can "buy" followers on eBay for less than a penny
each (TechCrunch
1/31/10
), which begs the question: is everything for sale or, do you get
what you pay for?

Do you see attribution models
playing a role in social marketing? 

Jim: Absolutely. Marketers need to know if they should invest 30% of their
social budget in blogs and 10% in tweets and 50% in Facebook or 10% in blogs
and 50% in tweets and 30% in Facebook. Without attribution, there's only
deciding by guess, by God and by gut feel. Oh – and by last year's (now
useless) results of course.

Bill:
The best marketers, and many Nielsen clients, are taking an enterprise approach
to measuring social media, recognizing that near real-time,
voice-of-the-consumer affects all aspects of the business from customer service
to marketing to public relations and corporate communications to design,
quality and the retail experience.  Also, the innovative companies are
working hard to evaluate social media as they do other types of media.
 This is how our best clients leverage the insights available via the analysis
of social media.

Carla: Yes.
While certain industry segments (Financial Services, Travel, Consumer
Electronics, Telecom) are embracing the use of attribution models as a way to
measure influential points across advertising channels, the most accurate
results have been when attribution models have been applied to the search
industry. As these models develop and gain greater adoption, their role in
accurately measuring social media will also increase.

Via the audience (Faisal Laljee): With social media,
messages and content can spread many degrees from the source. How far can and
should marketers track these ripple effects?  

Jim:  As
far as you can – within reason. If the seventh forward or retweet brings in
buckets of money, that is very valuable to know… unless it costs a ridiculous
amount of cash to find out.

Bill: The dispersion of social media about your company,
products, executives, issues, etc should be measured continually as it ebbs and
flows from one network to the next, from one influencer to the next.  At
Nielsen, we’ve seen social media catastrophes start with a decidedly
non-influential consumer and then get picked up by an influential blogger or
other influencer. It’s important to measure the message and understand
which ones are likely to go viral and the circumstances under which issues tend
to go viral.  

Carla:
Depending on the marketers' objectives, and the type of audience they are
trying to reach (early adopters, critics, commenters, etc.) the level of
analysis will vary.  Messages and content can spread widely, and often
even beyond the proverbial six degrees.  At Six Apart, we think all
degrees should be measured because, while some content can dissipate the
further from the source it travels, other content can gain momentum as it
ripples away from the source. Understanding how and when content dissipates and
which forms of content gain momentum can deliver valuable insights to inform a
marketer’s publishing strategy.  Ultimately, this insight can improve
social media effectiveness. Reference Altimeter's framework by Charlene Li and
the value assigned to each type of contributor in their "Engagement
Pyramid."

What needs more improvement right
now, the technological tools that marketers use, or the way marketers and
agencies use them and analyze the information? 

Jim: We have far more technology and waaaaay more data than we know what to do
with. What we need is technology that can identify what we should be looking at
rather than throwing more reports at us.

Bill:
The technology is out there but there is a lag in standard Key Performance
Indicators and a common currency that companies across industries can use and
leverage.

Carla: Both.
The technology and tools are still somewhat nascent, as are the metrics and
approaches for utilizing them.  The more marketers and agencies can
embrace a "test and learn" approach to social media measurement, the
more our industry can adopt measurement standards and best practices.

Via Twitter (smcatl): When is
predictive analysis going to become part of social media return on investment
(ROI) considerations?
 

Jim: ROI is a backward-looking concept and predictive
analytics is a forward-looking technique. Predictive methods are being applied
by scrappy start-ups as we speak. Stay tuned!!

Bill: Predictive analysis is already part of social media
ROI but it’s more of a niche metric because it is difficult to calculate and
the data needed to calculate isn’t always easy to get.  

Carla: See
above re: emerging metrics – "test and learn." All in good time!