HCF turns to crowd-sourcing in innovative customer retention initiative
Sydney, 12 March 2014 - In a first for Australia’s insurance industry, HCF, the largest not-for-profit health insurer, has sought ideas from the world’s top data scientists in an innovative crowd-sourcing initiative designed to enhance - HCF’s high customer retention rate.
Through US-based Kaggle, the world’s largest online community of data scientists, HCF last year invited 300 data specialists to submit ideas for a model that would allow HCF to better predict member behaviour and, particularly, the profile of members most likely to let their policies lapse.
After a strong response from the Kaggle community, HCF has selected three submissions for closer examination and analysis. The teams behind the winning models – comprising scientists located in Brazil, Singapore, Russia, the United States and Hungary - will share in a pool of prize money offered by HCF.
HCF expects the initiative will lead to a workable algorithm that will enhance HCF’s customer retention rate - already among the highest in the industry - and deliver richer insight into member behaviour.
HCF Chief Customer Officer , Stephen Nugent , said the Kaggle initiative gave HCF timely and cost-effective access to the best in expert global thinking on consumer predictive behaviour, which would ultimately lead to improved products and services for HCF members.
“As a customer-driven organisation, it’s critical that we understand as much as we can about what is important to our members and, in particular, what circumstances might prompt a member to leave HCF. Despite our already high retention rate we are always looking for ways to improve. Through Kaggle we now have access to a number of data management models that are customised to HCF and sharpen our understanding,” Mr Nugent said.
“We’re looking forward now to examining the winning submissions in more detail before building into HCF’s systems a predictive algorithm that allows us, among other things, to tailor our health cover more closely to member needs.”
In the Kaggle process, which was facilitated by Deloitte, the identity of HCF was kept hidden from the participating scientists invited to submit ideas. Membership data provided to participants was also de-personalised to protect the privacy of individual members.
Commenting on the initiative, CEO of Kaggle, Anthony Goldbloom, said: "We are very pleased with the results of HCF’s ‘As the World Churns’ competition, facilitated by Deloitte. This was a Kaggle Masters competition, reserved for the top 0.3% of the Kaggle community to gather and test their mettle against each other. Some 46 participants made more than 1800 entries for this challenge, using this large data set to create some very strong models."
