Customer intelligence in insurance

Identify and target customers with a second car


An online insurer would like to improve its marketing offer: It wants to propose to its customers a personalized offer concerning insurance for second family car.
The company at the beginning of the project had no information on the second car.
It has a limited client information (Last name / First name / Address / Email / Age / Profession). Furthermore it had a low representative sample of the population (5000 people).


Octopeek’s approach was based on the use of external and Open Data. We used phone book, social media, OpenStreetMap, and other sources.
The insurer’s data was enriched and that allowed the modeling of an algorithm.
The algorithm considers, among other things, population density, housing characteristics, income level, Socio-Professional Category, and proximity to public transportation.
score and a classification are then attributed to the initial sample.
This step makes it possible to move from an unsupervised machine learning model to a supervised one.

Then it was possible to extend the samples, to enrich the customer data further, to determine new correlations between their customers and an owner of more than one vehicle.
Samples are expanded and customer data is further enriched.
The final dataset obtained was composed of 90% new data and 10% of the insurer’s initial data.
The reliability of the algorithm is ultimately improved.
The company then created a scoring model that is added to the company CRM, that gives a probability that the current client has a second car.
Sales and Service process changes were implemented to require agents to ask high probability clients about the insurance for their second cars. It resulted in a 30% increase in new policies in 3 months.

Why Octopeek?

The insurance industry needs to move from ad-hoc experimentation with AI to enable everybody within the organization to use it to improve both operational efficiency and deliver business value. Octopeek aims to be the platform that democratizes AI in the enterprise: it makes possible the dynamic creation and provisioning of full-fledged enterprise AI applications, customizable by business analysts.


Improving vehicle quality

Exploitation of customer feedback during garage repairs


Dramatic churn reduction in insurance policies

Detect and prevent churn from unsatisfied customers


Telecom quality of service

Improve the quality, increase the renewal rate, reduce the bad payers


Credit scoring disruption

Time to market: Credit scoring for Gen Z


Process automation in investment banking

Automation of subscription warrant by OCR