Background:A US based telecom client that provides wireless service to customers wants to address customer concerns expressed in social media platforms like Facebook, Twitter, Blogs, Company Forums. This client uses Attensity 360 & Radian6 as the listening platforms to capture data posted on various social media platforms. The objective of this project was to accurately classify customer posts into sentiments- Positive Feedback, Negative Feedback (Actionable & Non Actionable)
Approach: AA series of simulation based key word/words algorithm was built in a statistical software. The accuracy of the algorithm was tested on the manual classification of the posts.
Impact: This project helped the client to accurately identify both direct & indirect mentions. The client is on able to reach out to customers with concerns more effectively than before. This helped increase customer loyalty, resulting in the increase of CSAT scores.
Social Media - A platform to sell, solve & communicate