Sentiments are complicated. Analyzing sentiments, even more so. The last few years have seen sentiment analytics becoming a critical component of customer feedback strategies for companies of all sizes.
Sentiment is easier to translate and analyse than it is to express. Sentiment analytics, also referred to as opinion mining, is a technique to abstract the underlying sentiment from textual data. Usually, this is customer feedback data flowing in from multiple channels such as email, call center logs, social media posts and even letters (if anyone writes those anymore). The idea is to understand not only the nature of the feedback but also derive context out of it. In that, sentiment analytics is both an art and a science.
Sentiment analytics uses a combination of Natural Language Processing (NLP), Machine Learning (ML) and deep text analytics to bring out the nuances hidden in text. However, it still has a long way to go to be perfect.
The complexity of spoken language makes it difficult if not impossible to derive sentiment accurately every time. Teaching a machine to differentiate tonality, discount grammatical errors, cultural adaptions, slang, consider rhetoric such as irony or sarcasm etc… is difficult.
Consider the following statements:
- “You want the customer to call you? Great!”
- “Your agents are simply brilliant. Not!”
Both the above statements are made up of words which if considered separately have a certain positive sentiment associated with it. But each complete sentence is filled with sarcasm. Existing sentiment analytics tools are not equipped to identify the true sentiment from such examples every time.
This is where sentiment analytics specific to industry lines can play a key role. In the insurance industry, sentiment analytics can be used in a multitude of ways that directly impact business. For instance, sentiment analytics can help insurers understand:
- The reason sales are falling in a particular zone
- Sudden demand for a product type
- Spike in customer complaints
The possibilities are endless.
Sentiment Analytics from Aureus
Aureus has worked on tons of insurance customer feedback data. Our sentiment analytics engine is driven by algorithms and data dictionaries fine tuned to work best for insurance. The engine is self-learning and is rule-based to help abstract context. Moreover, the users can suppress keywords that may not be relevant or club words that are similar in meaning but differ in spellings.
“Sentimeter” is an Aureus proprietary statistical model which can help insurers understand the true sentiment of any customer, at any given point in time. It is derived from all of a customer’s interactions and transactions with the insurer since the beginning of her contract. Sentimeter combined with Customer OneView provides sentiment analytics at an individual transaction level and at the customer level.
Sentiment and Business Growth
Sentiment analytics has emerged a very important business lever for many Aureus customers.
Some other cases where sentiment analytics can have a direct impact on business could be:
- Improvement in persistency / retention rates
- More efficient cross-sell
- New business growth by word of mouth promotion
Sentiment analytics can be a powerful tool if leveraged at the earliest, even if it begins with a small set of customers.