It's all about the numbers.

Our informational (we hope) and slightly irreverent (we think) take on all things analytics in the world of digital marketing and technology.
Posts I Like
Who I Follow
Posts tagged "social listening"

Not to be sarcastic but from having used sentiment engines extensively and having noticed the same kind of trends, we’d bet that the dominance of “neutral” ratings is more due to the limitations of the NLP (Natural Language Processing) algorithms generating the sentiment ratings than anything else.


"The overwhelming neutrality of posts is an indicator that people are increasingly relying on social media for customer service. Most of the neutral company mentions were from people seeking assistance and asking questions." 

Vast majority of company mentions are neutral messages | eMarketer 

In today’s series about “marketers have a lot of room to improve”, we switch to social media marketers: less than 5% of time spent on measurement? 12% of time spent on strategy? Ouch!

Another reason why brands should not be afraid to use social media as a customer service tool.

Finally…we have a majority of marketers using social media analytics!

To detect the mood about a stock on Twitter, Bloomberg relies on sentiment analysis tools that the company developed in-house. [Head of product at Bloomberg News] Rooney says that tasks like machine learning, auto-translation and natural language are all core competencies for an information company like Bloomberg, which has 3,000 engineers.
One of the most interesting uses of Big (unstructured) Data data we have seen in a while. Move over necessity, greed is the mother of invention. (from: Traders turn to Twitter for market news. Now they can measure mood, too — Tech News and Analysis)

These stats demonstrate, again, why it is so crucial for marketers to implement social listening and social media customer service strategies…and follow through on them.

'Nielsen has done a lot of research as to whether social media is helping people to tune in to TV,' he says, noting that a show's traction on Twitter and Tumblr is starting to affect how networks pitch shows to the all-important advertisers. 'They're no longer just including Nielsen ratings. They're also including social TV data.'

Interesting impact of social media analytics.

The Few, The Fervent: Fans Of ‘Supernatural’ Redefine TV Success | NPR 

(via unionmetrics)

A long and very geeky read but incredibly interesting and definitely written for the layman (i.e., us).  We have always been intrigued by NLP and how bad it is at capturing sentiment accurately in social media’s unstructured data.  This article explains why and how to get to a working semantic web.  Fascinating.

When we say it’s positive, the machine about 21% of the time says it’s negative. That can cause some problems in our understanding of how buzz impacts sales.
Another data point in the risks and issues associated with putting too much trust into NLP (Natural Language Processing) as a measuring tool for social media sentiment. (from: Coca-Cola Sees No Sales Impact From Online Buzz; Says Digital Display As Effective As TV)

Truly fascinating stuff that could revolutionize social listening and sentiment measurement.  It’s about time we move away from NLP (Natural Language Processing).


It is now possible to identify the meaning of words with multiple meanings, without using their semantic context


Two Brazilian physicists have now devised a method to automatically elucidate the meaning of words with several senses, based solely on their patterns of connectivity with nearby…

(via productivegossip)