96%of people think that customer service is a key factor to build a strong brand loyalty.
50%of people believe that companies should make decisions based on consumer feedback.
25%of users tend to replace a service or product after a single negative experience.
Therefore, the success of the companies depends not only on the service or product offered but also on the customer experience (CX) of using them.
Analyzing customer feedback is important to ensure that companies meet customer expectations for their products or services. The goal is to constantly improve the quality of products or services based on user experience so that companies can expect more returning regular customers and attracting new ones.
ChallengeA simple way to check customer satisfaction is through pre-defined question and answer forms.
Companies typically use the Net Promoter Score (NPS) metric to measure the loyalty of a firm's customer relationships. NPS is based on a single question whose answer ranges from 0 to 10 - how likely is it that you would recommend this company to a friend or colleague? In product reviews, the NPS normally ranges from 1 to 5 stars.
Although NPS is effective in estimating customer loyalty, it is not possible to find out the reasons that led to the calculated value. Even with extra questions it can be difficult to interpret customer feedback without a natural language report, which in turn lacks an automatic sentiment analysis of the texts.
OpportunitySentiment Analysis allows to enrich NPS's representativeness by automatically correlating other e-commerce CX indicators such as interactions, purchases, recommendations, reviews, and ratings.
The ease of extracting large volumes of relevant real-time customer sentiment data from social networks such as Facebook and Instagram makes social networking a strategic tool to support e-commerce.
For the project, we used the most modern sentiment analysis to evaluate consumer reviews automatically extracted from textual comments. EAI approach is based on the following points that assist companies to interpret their customers' feedback and solve such problems: variability of customer profiles, tracking customer sentiment over time, inclusion of context reviews, semantic text enrichment, and text normalization.