What is Customer Lifetime Value ?

In contractual insurance business model, Customer Lifetime Value (CLV) is the single most important metric for understanding our customers.  In insurance, the importance of CLV is similar to the credit score in bank and credit card companies. CLV helps us to make important business decisions about sales, marketing, product development, and customer support. For instance:

  • Marketing: How much should I spend to acquire a customer?
  • Product: How can I offer products and services tailored for my best customers?
  • Customer Support: How much should I spend to service and retain a customer?
  • Sales: We type of customer should sales reps spend the most effort to target for?

The heart of CLV is the notion of the retention rate, or retention probability.  An important managerial task is to take a series of past retention numbers for a given group of customers and project them into the future to make more accurate predictions about customer tenure, lifetime value, and so on. A defining characteristic of contractual business setting is that the departure of a customer is observed, called customer attrition. As we seek to understand the difference of customer behaviors in attrition and retention, we created machine learning models by analyzing data about millions of customers and their past retention / attrition numbers.  The data contains information about auto, property and life policies (C360), customer demographics (Acixom), customer segmentation (LifeScape), customer billing process and customer journey map.  Once the model is trained and optimized, it is able to predict the retention probability of a customer in any future tenure with the firm, from which we can aggregate and predict the expected tenure of a customer over a period of time.

The expected tenure, predicted by machine learning models trained on billion rows of data, is then used to multiply with the expected customer value to calculate the CLV of our customers. The CLV can be widely used in lots of decision making process, and can have a dramatic impact throughout our business.



Chatbots: A new frontline to empower convenient intelligence for insurance industry


We human beings have been shackled to keyboards and monitors for more than forty years. Now it is time for a change. We need a new intelligent and universal mean of interaction that can hold voice or text conversations to link humans and machines, and to connect customers and companies.  In this new wave of technology, conversation is the new interface, and chatbots are the new applications. The power of the natural language processor, software that process and parse human language, enable Chabots to create a simple and universal means of interacting with technology.

Comprehending human conversation is no easy task for a computer, and companies all over the world have sunk hundreds of millions of dollars into the effort. In our lab, we have successfully built a series of product assistant chatbots to help our agents querying and searching product knowledge extracted from product reference manual website. Our chatbots is based on Google chatbot platform api.ai, and integrates text/voice input from Slack, Skype, Facebook Messenger, Mobile Applications and Amazon Alexa.  Our knowledge base was constructed by scraping sharepoint website, and transforming scrapped unstructured text to knowledge base.  Our knowledge query APIs were based on Elastic Search.


Build a service that never sleeps

Building Chatbots is an easy way to streamline that interaction by providing information to a customer faster and more efficiently than a customer service representative.

Never miss a meeting again

Even for processes not dependent upon customer service, chatbots can still be useful for any number of reasons. Just consider an administrative bot that schedules meetings for you. While it may seem like a mundane task that doesn’t need its own bot, this kind of practicality is incredibly useful and is a big time saver.

Access information in a more productive and consistent way

Have you thought how many time we waste each day on finding information in complicated computer systems or search through unorganized web content?  Even sometimes we know where the information is located, to ensure not missing the important updates, we have to check and confirm the correct version of content again and again? A well-designed Chatbot can save all these headaches.  By asking the correct questions, we can command chatbots to retrieve information efficiently from the consistent repositories.

Empower the convenient intelligence

Chatbots are the frontline of our enterprise level intelligent applications. Let’s Imagine an intriguing scene: one day our chatbots have integrated services such as recommender system, customer life time value scoring, insurance quoting API, telematics data analysis and real-time claim processing, etc., and are capable of delivering convenience to our customers at anytime and anyplace. We all know that day is not far away, and we are all excited to see its coming as a new dawn of revolutionized insurance industry.