More and more companies today are looking at the sentiment consumers are voicing about their brands in online forums, blogs, microblogs, etc.  There’s good reason for this as it has been shown that positive and negative word of mouth correlates with growth.  “Answering The Ultimate Question” is a book I’ve been reading recently makes a strong case for the positive word of mouth as a leading indicator for revenue growth.  What that means is a brand manager should be able to predict their revenue growth based upon what’s currently being said about their brand.

With the amount of data being generated by consumers out there, managers need tools to detect and analyze sentiment automatically.  The hardest thing about getting sentiment analysis right though is uncovering exactly what is being said by a piece of sentiment-rich text.  There’s a big difference between a customer saying they merely like a brand and saying that they love it.  What’s more is sentiment has many rich and nuanced dimensions that need to be teased apart to make it truly insightful.  One person posts on Twitter, “Old lady at Wal-Mart told me warm Dr. Pepper was delicious.”  Another comments, “I loved Dr. Pepper because of Dude.”  Aside from avoiding the common blunders of associating the sentiment with the wrong brand in the tweet, deep sentiment analysis needs to unpack the specific drivers of the sentiment.  In the first sentence, the sentiment for Dr. Pepper is positive because someone thinks it’s delicious.  Contrast that functional benefit with the rich nostalgia expressed in the second sentence where a Baby Boomer recalls the famous Dr. Pepper “Dude” in the “I’m a Pepper” TV commercial .  The best algorithms out there have to go deeper than sentiment detection.  They need to be true sentiment analyzers that can precisely identify sentiment, extract it, and classify it as a functional, emotional, or behavioral insight, etc..

In the coming year, I believe we’ll need sentiment analysis that can get to very fine levels of detail while keeping up with the enormous and growing volume of social media.  For instance, the “Ultimate Question” book I referenced earlier is all about whether a consumer likes your brand enough to recommend it.  Loose sentiment detection isn’t precise enough to answer this type of question because it lumps everything from the high-end of recommending to the low-end of merely liking into one positive bucket.  Managers will demand this type of precision as they look more and more to make critical business decisions based on insights derived from sentiment analysis.


NetBase Product Line

Premier social media analytics platform

Tailored platform for growing businesses

Expand your social platform with LexisNexis news media

Power of social analytics for your entire team

Customer experience analytics platform

AI, Image Analytics, Reporting Tools, APIs & more

Product configurations to meet all needs

Quid Product Line

Media analytics and market intelligence platform

Enrich your media analytics with social data

Media coverage for historical & real-time monitoring

AI algorithms, NLP, data sources, and visualization

Tailored, configurable solutions