With open-ended comments, customers get to tell us what they really care about and why without being limited to the questions we decided to ask. Using this data for making the best decisions, though, depends on the analysis. Here are some important considerations when selecting a text analytics platform:
Open-ended comments from sources such as social media, customer surveys, and your contact center can provide rich insights for both the brand and its locations. Executives can see and react to the macro-level brand relevant commentary that will help them shape strategy, products, merchandising, customer experiences, and pricing. Location-level operators and franchisees can use commentary to drive and focus training, empowering teams to execute both brand and operational excellence.
Using this data for making the best decisions, though, depends on the analysis. Here are some important considerations when selecting a text analytics platform:
- Natural Language Processing (NLP) Algorithm: Choose an engine with excellent NLP capabilities, like IBM Watson’s Alchemy program. This algorithm allows computers to ‘read’ the text that a human has written. Any NLP engine should provide you with categories, more refined categories, entities, word patterns, keywords, and sentiment.
- Dataset relevance: Make sure that the questions you want to ask are targeted towards a relevant dataset. As an extreme example: Understanding brand sentiment from a contact center stream is going to give a heavily biased outcome to negative sentiment as many customers use this channel to complain—not provide compliments.
- Dataset sources: Any datastream with open-ended comments is fair game. Consider using social posts, customer satisfaction surveys, contact center comments, and even mystery shopping or audit annotations.
- Analytics: Online platforms should present categories of data, trending, and be able to pinpoint changes. Analytics teams can take these analyses one step further by using text categories in two ways: 1) As predictors of satisfaction and loyalty and 2) To show the differences in and commentary between different groups. For example, a small business entrepreneur pays attention to and comments very differently on their cell phone purchase experience than a personal user. These differences in conversations can give you keen insights into how to address the needs of various customer segments.
Using our text analytics platform capabilities, sentiment and keywords can be tracked in near real-time giving you the ability to track emerging trends—and as with all data, feeds into our KnowledgeForce® platform to provide you a single cross-datastream analysis of what your customers say.
We’d be pleased to provide you with a demonstration.