Our client runs a price comparison platform where, among other things, visitors can obtain rates from more than a hundred auto insurance providers. But primarily, users have to complete the online form with around 50 fields, which takes quite a lot of time and discourages many potential customers.
To engage more car owners at the start and increase the conversion rate, the client developed a quote predicting widget with a reduced form to be placed on third-party websites. Based on just a few parameters, the tool forecasts an average cost for insurance and, if the users are satisfied with it, redirects them to the main platform. AltexSoft’s data science team created a machine learning model behind the quote predictor.
The project’s scope was 2 man-months. It was completed by one Data Scientist.
The technology stack of the project included Python, Pandas and scikit-learn libraries, Flask microservices, and Docker (for API building and integration). The machine learning approach included the LightGBM algorithm.
Cooperation between the client and AltexSoft is ongoing.