We have added a new capability inside the feedback product – Sentiment Analysis. We’ll programmatically classify all the incoming user feedback on your website as – positive, neutral, or negative in its sentiment.
The purpose is to give you an idea of the customer happiness quotient, simply by putting a time series of how the customer sentiment is changing over a period of time on your site. You can access this chart under Feedback > Stats section. Here’s a glimpse –
Also, each of your feedback threads in the dashboard will display its corresponding sentiment as detected by our self-learning algorithm. Underneath are samples of classification from our own customer feedback (notice the sentiment icon on the upper right corner for each of the feedback emails).
This is an experimental feature. And the program is an auto-learning one. It does a decent job and with time it will only get better. If you notice a thread being incorrectly classified, you can make our classification algorithm a bit better by taking the pain and re-classifying as shown underneath –
Under the Feedback > Stats section, we also give you an overall distribution of sentiments across all the feedback received on your website.
Hope you liked what you saw. Your feedback on this feature will help us improve.