SingaporeFinance and insurance2018 – ongoingNTUC Income
Machine learning audience segmentation for insurance marketing
NTUC Income cut customer acquisition cost by 40% after machine learning segmentation of its insurance audiences
Problem
Income's customer data sat in separate systems: the CRM database, email tools and website analytics. The marketing team spent months planning campaigns to find target audiences. The on-premises customer database also did not show who intended to buy a given policy.
Approach
Income worked with the Singapore agency Happy Marketer, which set up Google Analytics 360 and BigQuery to bring customer interactions together. Income's data scientists run machine learning models in BigQuery that split its 1.8 million customers into fine segments, such as likelihood to buy and preferred marketing channel. Optimize 360 then serves personalized messages to each segment. The team launched 10 use cases between 2018 and 2020.
Results
Takeaway
Income tied its marketing results to a propensity-to-buy model fed by joined-up customer data. The company did not say how it measured the acquisition cost change.
Categories
- Industry
- Finance and insurance
- Business function
- Marketing and sales
- Innovation type
- Process innovation
- AI technology
- Predictive analytics and ML
- Adoption
- Off-the-shelf tool or platform
- Evidence
- Vendor case study
Sources
Source summaries
Google Cloud's customer story describes how Singapore's insurance cooperative Income used BigQuery and Analytics 360 to score customers on their propensity to buy and personalize marketing. It reports a 40% lower customer acquisition cost and a 92% higher click-through rate, and says the team launched 10 use cases between 2018 and 2020. The page carries no date; the Internet Archive first saved it on March 26, 2021.
Full details are in the original sources.
