Get “Sentimental” for Valentine’s Day

By: Lisa Sigler

February 12, 2016

Sentiment Analysis
Social Customer Care
Social Engagement
social listening
social media
text analytics

Love may be in the air this time of year—but you can’t just assume your customers have tender feelings for you. One way to know for sure is through sentiment analysis of your customer feedback.

Customer sentiment analysis tells you how positively or negatively your customers feel toward you and toward topics that relate to your brand. These topics might include your products, your service, your location, your advertisements, or even your competitors.

Clarabridge performs customer sentiment analysis in tandem with text analytics. If you do it right, with sentiment analysis that understands the nuances of grammar, your customer sentiment analysis will show you if your customers are feeling the love.

Does he like you, or does he like you-like you?
Is it a passing crush, or are your customers falling for you hard? Clarabridge uses an 11-point sentiment scale for a really specific view of how positive or negative your customer feedback text is.

This provides significantly greater accuracy than a 3-point “positive/neutral/negative” rating found in many sentiment analysis models. This accuracy is good if you want to take action on your feedback—since we all know that there is a big difference between a good meal (+1) and the best food you’ve ever eaten (+5).

She loves me. She loves me not.
The Clarabridge sentiment analysis model recognizes the incredible power of the word “not.” Referred to as negation, adding the word “not” changes the positive sentiment of a word like “happy” or “delicious” to something negative (and makes negative phrases positive, too).

Similarly, conditional sentiment refers to statements that may include positive words but are neutralized by phrases like “kind of” and “sort of.” If you want an accurate view of your customer sentiment, you need to understand these grammatical nuances.

At just the right time and place
Words can have different meanings depending on how and when they are used (as anyone who has ever had a lovers’ quarrel can agree). That’s why you need context-dependent sentiment, to recognize that in some contexts a phrase may be positive, while in other contexts it is negative. Clarabridge has developed industry- and situation-specific templates to automatically recognize if words are considered positive or negative for your specific needs.

You are too exceptional
Exception rules recognize grammatical constructions that consistently change the sentiment of a word or a phrase.
For example, the sentence “Roses are red,” would normally have a neutral sentiment value (a “0” on the 11-point scale). However, “Roses are too red” is negative. Adding “too” to a neutral becomes a negative so often that Clarabridge has a rule to account for it—along with 500 other rules in English alone.

Understanding how your customers feel about your brand is crucial if you want to take actions that matter. Maybe this Valentine’s Day is the time that you recommit to listening to your customers and taking their emotions into account in your business plans. Show them that you really care.

If this peek at customer sentiment analysis has sparked your interest, or if you really want to get to know us better, The Truth About Text Analytics and Sentiment Analysis
will let us take our relationship to the next level.


Lisa Sigler is Sr. Manager of Content Marketing at Clarabridge. For over 16 years, Lisa has used her writing and editorial skills to bring the value and benefits of technology to life. In her current role, she works to demonstrate Clarabridge’s position as thought leader and trailblazer in the Customer Experience Management market. Lisa holds a B.A. of English from Kent State University. Read more from Lisa on Twitter @siglerLis.