(^147) ■
SOCIAL ANALYTICS
about you. Your active listening program takes this one step further by connecting these
sources with the actual business impact of the conversations in which they are a part.
Knowing who is talking is an important part of understanding the meaning of
what is being said and then applying this in a useful manner within your business or
organization. Combining the sources of the conversation—especially when the sources
are actual (or potential) customers—with your listening data provides insights into how
you can evolve your product, how you can reshape the customer experience, and where
you and your competitors have points of relative vulnerability. As you consider specifi c
listening tools and listening programs, consider how the data provided facilitates con-
necting the source of the conversation along with the actual conversation and what it
means to your business processes that are driving the conversations in the fi rst place.
One aspect of listening—known as sentiment analysis—deserves specifi c men-
tion. Sentiment analysis is one of the meaning-related processes by which conversations
are categorized. It’s also one of the most talked about and one of the most troublesome.
Without belaboring the point, human communication is complex: Anecdotal examples
of the issues around sentiment analysis can be summed up in two words: “Dyson
sucks!” Is this a positive comment—perhaps an expression of joy by a new customer of
these amazing vacuum cleaners? Given the reputation for the Dyson brand, it’s more
than likely the case that this is a positive comment.
Sentiment analysis, important enough in its own right, is not an end in and of
itself, though many would love it if it were. How great it would be if instead of actively
listening all you needed to do was read a report and respond to eight negative posts
or send “thank you” notes to a dozen loyal fans. Unfortunately, there’s more to it. A
lot more. Like the tip of an iceberg, sentiment analysis—and more specifi cally track-
ing and trending sentiment—gives one an indication of what is happening below the
surface. Unlike the tip of an iceberg—which is a good indicator that there is more ice
underneath that is otherwise just like the ice you can see above the surface—in the
case of sentiment analysis, you generally have to go back to the original posts, to the
original context of the conversations, to sort out what is really happening: you may
fi nd more ice, or you may fi nd rocks. A strong negative comment may originate from a
dedicated fan, or a dedicated detractor. You must dig in and understand these kinds of
differences in addition to simply following the sentiment score.
TweetTone
Looking for meaning and emotion on Twitter? Check out TweetTone. It’s a clever application that
enables you to quickly look for different topics—all review a basic tonality analysis, all within
the context of the individual tweets that gave rise to the tone results. You’ll find TweetTone here:
http://tweettone.comTweetTone
Looking for meaning and emotion on Twitter? Check out TweetTone. It’s a clever application that
enables you to quickly look for different topics—all review a basic tonality analysis, all within
the context of the individual tweets that gave rise to the tone results. You’ll find TweetTone here:
http://tweettone.com