The ever-changing digitally savvy customer now has high expectations and digital transformation is a key priority for Insurance companies to keep customers engaged and to drive efficiencies.TechVantage has a solution portfolio for P&C insurers that makes use of customer data that Insurers have and leverages TechVantage’s man-machine model.
As part of the insurance claim process, customers are to upload a video testimonial to report the reason for the loss of the device they have insured. The requirement was to analyze the video and use non-verbal cues to detect any fraud indicators.
Video analytics using Python-based modules.
Experts opinion was used to define behavior exhibited when a person tends to lie.
Image and video analytics was used to detect those patterns and arrive on the probability of fraud.
Automation of this task enabled cost savings for the insurer. Reduction in fraud and a more efficient claims process
The biggest challenge faced by the insurance sector is that of fraud. The task here is to identify groups where users know each other, so as to reduce the propensity of a user to commit the fraud.
Groups are allocated to people on the platform based on data collected during the login process like people to whom they talk to, are connected via social media, share the same location etc.
Each individual gets a score based on which the user is assigned to a group.
An intelligent grouping mechanism which puts people knowing each other in the same group so as to minimize the tendency to commit fraud.
Client was launching a new product for smartphone insurance. A campaign was run across various channels (Fb/Google/LinkedIn/Agents/Ecommerce). The task was to measure the effectiveness of the campaign in order to understand the interest towards such a product.
Client conducted a survey based campaign which asked the users to fill few information along with their views towards the new product.
The responses were analyzed and customer segmentation models were built to understand the general profile of interested customers
Based on the responses most effective channel to reach different target clusters was identified.
It was identified that millennial men living in Jakarta were the most likely target population.
Facebook was identified as the most effective medium which generated maximum click-through-rates.
As part of digitization, the Insurance company wanted to create a chatbot for their product information. The chatbot should address queries on products listed on their platform.
An intent-entity based NLP solution was developed to address customer queries on insurance products.
The textual information contained on the website and FAQ section was used as the corpus.
A CNN model was used for training the bot.
A web based chatbot which could answer customer queries with an accuracy of 80+% was developed.
In cases where the bot could not answer, it would redirect the customer to the customer support executive.
Post implementation, number of customer support executives were reduced by 67%.
Average Customer Response Time decreased by 70%.