As an unabashed lover of data, I am thrilled to be living and working in our increasingly data-constructed world. One of the new types of data analysis that is eliciting strong emotional reactions these days is the sentiment analysis of the now directly digitized feedback from customers provided via online reviews and social networking, where word of mouth has essentially become word of data.
Sentiment analysis is often marketed as an essential component of understanding what your customers really think about your products and services. Although it can definitely be valuable, sentiment analysis can also suffer from what is known in psychology as negativity bias.
Customers, like all people, pay more attention to, and give more weight to, negative experiences.
If you doubt the sentiment of that statement, I welcome you to compare being complimented with being insulted. Which are you more likely to remember? Which are you more likely to tell people about?
Yes, self-aggrandizing people often brag about the compliments they received, as well as invent compliments they never received. And, of course, this aspect also skews sentiment analysis because false praise and fake five-star reviews arise as a result of the dark side of social media marketing.
But authentic customer feedback is frequently negative because of our negativity bias. For example, we only contact customer service when we have a negative experience with a product or service.
During my college days, I had a night job as a customer service representative, and I don’t remember ever hearing a customer say: “I just wanted to call and tell you that your product works really well and your service is absolutely fabulous!” However, I do fondly remember learning some new swear words, which was impressive since I grew up in a family that, paraphrasing a line from A Christmas Story, worked in profanity the way other artists might work in oils or clay, it was our true medium.
Now, this doesn’t mean the dark side of the Force is stronger, it simply means that we all have a natural tendency to focus more on the negative aspects, rather than on the positive aspects, of most situations, including data quality (e.g., although we often hear people complain about bad data quality, when was the last time you heard someone praise specific instances of good data quality?).
A common misconception about the far side of the Moon, which, since the Moon is tidally locked, is permanently turned away from the Earth, is that it is dark even though it receives approximately the same amount of sunlight as the near side. The Dark Side of the Moon actually means that it’s in a radio blackout in relation to transmitters on Earth. In other words, if you were trying to listen to a radio while you were walking on the dark side of the Moon, you wouldn’t be able to hear anything.
When you are listening for customer feedback using sentiment analysis, you will be able to hear it, but the Dark Side of the Mood might prevent you from hearing anything other than negative feedback.
post: Dator on the Carrots and Sticks of Sentiment Analysis