Customer Service by the Numbers

Posted by in Customer Service


One of the problems with online customer service is the lack of visual and vocal clues that help to understand the writer’s intent or feelings.  Words on a screen or in a text message can be difficult to understand without the emotion behind them.   Positive comments can be taken negatively, and vice versa, without clues to intent.

 

What if you could not only read customer comments but know the feeling and emotion behind them?  Well, now you can decipher the meaning behind the message. According to an article in Business Insider, “How Starbucks And Other Companies Use Complex Math Algorithms To Read Your Feelings Online,” a new technology called sentiment analysis uses complex mathematical algorithms to differentiate between positive and negative statements.    This breakthrough software can tell if a customer is complaining or is happy by determining how they use words in specific combinations.  The mathematical calculations humming away in the background “read” the emotional equations and let companies know which customers need some personal attention.

 

The software isn’t smart enough to do this alone.   Some positive words or phrases can represent negative feelings in certain combinations.  That’s where the programming comes in.  The software has to be taught to recognize positive and negative feelings based on the company’s language and phrasing.  Companies can customize the math to their specific needs. 

 

The beauty of the technology is that it can constantly monitor the huge volume of customer feedback on social media sites, such as Twitter, Facebook and thousands of blogs when they mention a company name, product name or specific keywords.  Starbucks uses it to analyze tweets that come in at 10 per second and pull out those that need a response.  A company would need legions of customer service reps to handle the volume of feedback available online, phone and in person.  This technology can scan hundreds of thousands of electronic messages and pull out those that need attention.

 

According to the article, The American Cancer Society is able to gather feedback on its various programs and events to see what participants liked and would like to see in the future.  The article stated that the ACS gets about 62,000 comments about their Relay for Life event.  They used sentiment analysis to find out what they could do to make the event even more effective and meaningful.  Non-profits can use the technology to stage more effective events to increase fund-raising and awareness. 

 

With so much commerce going online, the world is the marketplace.  A small, local company can have customers halfway around the world and around the corner.  This new sentiment analysis software can make it easy to find out if a customer in Singapore liked the latest shipment of Aunt Susie’s Lemon Curd by monitoring social media.  Most unhappy customers don’t complain.  They just don’t buy.  This new technology will take advantage of social media’s obsession with sharing everything to find out how they’re doing in a global market. 

 

Photo Source:  Freedigitalphotos.net

Comment

Become a member to take advantage of more features, like commenting and voting.

  • Mary Nestor-Harper
    Mary Nestor-Harper
    I never knew there were so many types of software that can predict a customer's emotions.  Such helpful tools for service reps to do a better job.  Thanks for the comments.Mary
  • Mark G
    Mark G
    This would really help companies improve their customer service because I find not only in person but over the phone, or in person, many people are,  that shouldn't be working in customer service business. Most of them don't show the true empathy, ask the right questions and exacerbate an existing bad condition.
  • Dorothea L
    Dorothea L
    This could be very helpful to determine properly what the customer is trying to get across. & emotions is all about customer service this is the saviors of a company.
  • Michael R
    Michael R
    Actually, there are interactive voice response unities that are capable of identifying service escalation problems - language independent! They consider db level increases, tone emphasis and other criteria. These systems are now deployed.
  • Mary Nestor-Harper
    Mary Nestor-Harper
    Thanks, Sandra.  This technology is a great tool for customer service reps that may not be able to discern a customer's true feelings.  It can help companies give better responses to complaints.  Glad you liked the article.
  • Sandra W
    Sandra W
    I worked in customer service for an international lighting and audio company for the entertainment industries. I think this software is AMAZING. It is amazing in the last ten years since I worked how many advancements there have been. Truly, this would really help companies improve their customer service because I find not only in person but over the phone many people that shouldn't be working in c.s. but they are not told and even if you like the product you would be apt to go somewhere else to avoid a person with bad c.s skills. Great technology and article!

Jobs to Watch