Mobile customers frequently buy insurance at the time they buy new phones. In case a phone is broken, the customer can claim insurance which involves raising a request with images of the broken phone. Claim verification involves detecting damages/cracks, IMEI number and tempered glass presence. Image classification models can classify images of screens and tell us whether they are broken or damaged or not. Object detection and OCR models can read an IMEI number present on the screen for identification purposes. Automating the process of detecting cracked screens means customers can upload images of their phones from anywhere and you can automatically review each claim and initiate refunds or replacements without any hassle, enhancing the efficiency of your organization, reducing costs of manual reviewers and increasing customer satisfaction.
Traditionally, the process of making an insurance claim first involves taking images of the phone by the user. The user then sends these images to his/her insurance company. The company then sends these images for manual review, the time taken for which depends on the image quality and filtering out the fraud images. This process, if the image turns out to be good takes a week. If the image isn’t fit for inspection, the user is informed after one week. He/she is expected to resend the images to the company so the company can make sure their claims processing procedure is accurate. Once the claim is approved, the user is put in communication with the finance team of the company to carry forward the procedure.
Inspection with Nanonets
With Nanonets, the user gets immediate feedback on if the images he/she has taken are good for inspection or not. The will know about it in a second following which he can update images and move to the next step. These images are now uploaded for inspection, which is also automated with state-of-the-art machine learning models which provide accuracy at least as high as human reviewers. Besides this, the process of inspection is cut down from at least a week to less than a day. Apart from increasing customer satisfaction and convenience, the company also saves a lot of money by cutting on the cost of manual reviewers. The claims, approved or disapproved are added to the insurance company's database for future use.
The Nanonets API
Our API provides high speeds and great accuracy, enables fraud detection and drives automation for insurance companies. The Nanonets API can help the insurance companies automate the process of
- Finding out if the image adheres to specifications like the high resolution, full phone being visible, upright orientation of the phone, no reflection of flash on the screen, etc. and providing instant feedback to customers.
- Inspecting images and detecting:
- The IMEI number of the phone
- If the screen is cracked or not and with what severity
- If the phone uses a tempered glass or not.
The Nanonets Impact
We were able to provide 80% automation for the total volume for the company. This led to a more than 60% reduction in costs along with an 80% reduction in the time taken for the process to reach completion excluding error handling. The Nanonets API was able to deliver accuracy comparable to humans. The human error rate was found to be 0.5% and the Nanonets API performed with a 0.77% error rate.