Client
Hedvig
Tech stack
Google Cloud
Solution
Marketing Analytics
Service
AI + Machine Learning
Boosting Hedvig’s Profit by 20% with Member Lifetime Value
Hedvig is a Swedish insurance company that is disrupting the insurance industry by providing a reliable and personalized service to handle customers and claims within minutes. In today’s digital-first world, understanding and optimizing customer lifetime value (CLV) is more important than ever. But Hedvig is doing it differently—by adopting a fresh approach: Member Lifetime Value (MLV). Hedvig collaborated with Datatonic to make this possible using best practices on Google Cloud, allowing them to optimize customer acquisition costs and boost revenue.
Our impact
- Increased profit by an estimated ~15-20% using MLV data to reduce customer churn and optimize customer acquisition costs
- Optimized decision-making by creating a powerful MLV model for predictive model to estimate future customer lifetime value for three, six and twelve months periods
- Enabled Hedvig to improve its targeted marketing campaigns with MLV and Google Analytics data
The challenge
Since 2018, Hedvig has been working hard to create an insurance company that is based on trust, is easy to use and designed for how people live today.
Hedvig wanted to better understand its members so that it could optimize decision-making around advertising, marketing, and pricing.
To do this, Hedvig wanted to develop a Machine Learning model to predict Customer Lifetime Value (CLV) using services on Google Cloud. By predicting CLV, Hedvig can segment customers and execute more marketing campaigns with a much higher precision, amongst other activities.
Hedvig wanted to take this further by combining CLV with its member-focused approach. Member Lifetime Value (MLV) places an emphasis on people being seen as members where interactions are encouraged to ensure members are able to benefit from their insurance and feel supported.
“We knew that the future value of each member was different, but lacked the tools to act on it. With Datatonic we have created the tool, MLV, to act in real time, improving both acquisition efficiency and member experience” – Simon Skogman, Acquisition & Monetization Lead, Hedvig
Our solution
With a clear goal in mind, Datatonic collaborated with Hedvig to develop powerful predictive Machine Learning models to deliver insights about Member Lifetime Value:
- Provided evaluation results of benchmarking three different Machine Learning models before starting.
- Developed a set of Machine Learning models capable of estimating Customer Lifetime Value, for three, six and twelve months periods.
- Created two pipelines, for training and inference, composed of data ingestion, data pre-processing, model training, model evaluation, and model deployment components.
- Supported all operations on SQL, including ML models using BigQuery ML
- Used Infrastructure as Code (IaC) to provision the resources on Google Cloud needed to deploy and execute the pipelines.
- Developed CI/CD pipelines to automate the deployment of code.
This new Member Lifetime Value capability enables Hedvig to improve its marketing efficiency and reward its most loyal customers with low and fair insurance prices.
Find out more by watching our presentation from Google Cloud Summit Nordics.
