AWS Partner Network (APN) Blog

Revolutionizing User Experience on Claim Submissions Using WhatsApp Chatbots

By Clevester Teo, Partner Solutions Architect – AWS
By Abhijit Kalita, Sr. AI/ML Evangelist – AWS
By Kelvin Yap, Lead Architect – Axrail

Axrail-AWS-Partners-2023
Axrail
Axrail-APN-Blog-CTA-2023

In today’s fast-paced world, social media has become an essential part of our lives. People use social media platforms to stay connected with friends and family, share experiences, and express themselves. Traditionally, filing a claim with an insurance company has been a time-consuming and frustrating process. Customers have had to fill out lengthy forms, provide documentation, and wait for a response.

But did you know social media can simplify a claim process for expenses such as medical bills? As an intuitive and functional tool, WhatsApp chatbots might be the solution to address such a process. WhatsApp chatbots are conversational AI-powered bots, automating conversations with humans that take place on WhatsApp.

Chatbot technology is rapidly revolutionizing customer experiences, providing businesses with a way to offer immediate and curated responses without allocating additional manpower and resources.

WhatsApp chatbots are incredibly easy to use. Simply send a picture for submission or ask a question and the chatbot will respond with accurate answers in no time at all, making it much faster for businesses to handle large quantities of inquiries in a short period of time.

In this post, we’ll showcase how using social media like WhatsApp in conjunction with Amazon Textract can make the claim process more streamlined and efficient, benefiting both customers as well as claim processing agencies.

This solution is developed by Axrail, an AWS Advanced Tier Services Partner that offers cloud services, ecommerce platforms and automation, artificial intelligence (AI), and analytics.

Solution Overview

The objective of this solution is to simplify the claim process for any kind of expense. Users will use WhatsApp to start the claim process, and they can either submit a new claim or check claim status in WhatsApp.

For new claims, users will need to upload the bill in one of the Amazon Textract-supported formats. Amazon Textract is a cloud-based service that uses machine learning (ML) to automatically extract text and data from scanned documents. It can recognize text in images, tables, and forms, making it an efficient tool for processing claims.

Once the bill is uploaded, the Amazon Textract Analyze Expense API—which is custom built for invoices and receipts—will be invoked to extract the key-value pairs such as Invoice Receipt Date, Invoice Receipt ID, Customer Number, Vendor Name, and Receiver Name.

There are two ways to call the Analyze Expense API: synchronous and asynchronous calls. If you have a single receipt to be processed, then you’d be using the synchronous API to get faster response. If you have large number of receipts, you can use the asynchronous API to queue and process the invoices sequentially.

The user has the option to confirm or edit the content. Once confirmed, the claim will be submitted to the backend system and the backend database will be updated automatically. This saves time and reduces the risk of errors while ensuring the claims are processed more quickly and accurately.

Axrail-Textract-Claims-1.1

Figure 1 – Customer journey for the claim submission flow.

By integrating WhatsApp with Amazon Textract, insurance companies can offer their customers a faster and more convenient way to file claims. Instead of filling out a lengthy form or submitting it via traditional channels like emails, customers can simply take a photo of the claim form and send it to the insurance company via WhatsApp.

Reference Architecture

Let’s look at the architecture diagram:

Axrail-Textract-Claims-2

Figure 2 – Axrail’s WhatsApp-integrated customer service platform.

  • To enable users to easily connect with this solution, insurance companies can distribute their WhatsApp Business Profile by sharing a QR code or contact number.
  • When users need to submit a claim, they’ll be required to send any type of message to this account to initiate a 24-hours customer service window. This allows the solution to send the message back to users.
  • Integrating WhatsApp Business with Amazon API Gateway involves creating an HTTP endpoint to receive incoming messages via a configured webhook, which then triggers AWS Lambda to handle the received messages, allowing insurance companies to easily automate communication with their users.
  • The API now has a Lambda function as the backend. Multiple Lambda functions are orchestrated by AWS Step Functions, a serverless orchestration service that provides a flexible and scalable way to coordinate the functions.
  • In a typical workflow, the first Lambda function in the AWS Step Functions orchestration process would receive the uploaded receipt and store it in an Amazon Simple Storage Service (Amazon S3) bucket, followed by a call to Amazon Textract’s AnalyzeExpense API, which utilizes ML to extract data from the receipt for further processing without the need for any templates or configuration.
  • In the event of a poor quality image or incorrect file format, the if-else flow in AWS Step Functions invokes a second Lambda function in the workflow to prompt the user to resubmit the bill with a higher quality image in the correct format for accurate claim analysis. Amazon Textract’s AnalyzeExpense API provides a confidence score to indicate the likelihood of accurate text extraction.
  • After the expense data has been extracted, the user is presented with a preview of the extracted text from the uploaded bill, including all of the fields that were mapped during the expense analysis process, and is given the ability to edit the information using a WhatsApp call to action (CTA) button if there are any errors or inaccuracies in the extracted data.
  • Once the user has reviewed and confirmed the extracted expense data, the final text is stored in an Amazon DynamoDB table for efficient retrieval and processing, while the data is batched and transferred to the relevant claim system for financial processing. This enables the seamless and automated processing of expense claims for increased efficiency and productivity.

Conclusion

Using social media platforms like WhatsApp in conjunction with Amazon Textract can significantly simplify the claim process for both customers and insurance agencies. It allows for a faster and more transparent claims process, leading to improved customer satisfaction and loyalty.

In addition, using WhatsApp as a communication channel allows insurance companies to easily communicate with customers throughout the claim process. This two-way communication helps build trust and transparency between the customer and the insurance company, leading to higher customer satisfaction and loyalty.

This solution from Axrail is fully serverless with high availability and scalability that can easily be customized for similar use cases. Connect with Axrail to learn more about this WhatsApp-integrated customer service platform and accelerate your time to market.

.
Axrail-APN-Blog-Connect-2023
.


Axrail – AWS Partner Spotlight

Axrail is an AWS Partner that provides outcome-based digital transformation through analytics and app modernization using cloud-native capabilities.

Contact Axrail | Partner Overview