From competition to call centre in five months: Zimnat sets a new record for putting Zindi models into practice

Image for post
Image for post

“We asked Zindi to give us a way to predict when a customer would ‘churn’ or leave as a paying customer,” says Zimnat’s Chief Digital Officer, Oswin Zulu. “We built the winning models into a dashboard with a customer profile and the likelihood of them leaving the business, and that dashboard is now given to our customer care agents.”

The Zimnat Insurance Assurance Challenge on Zindi attracted just under 300 Zindi users from all over the world, and in 60 hours in May produced winning solutions from users in Nigeria, Tunisia and India. Just five months later, Zulu and his team had deployed their customer care agents’ dashboard using those models.

In providing their agents with direction on who may churn and possible interventions to help prevent that, the machine learning solution produced by Zindi has had a rapid and direct impact on Zimnat’s core business — providing great insurance products to Zimbabweans.

Digital transformation in Africa

Zimnat is one of Zimbabwe’s leading insurance companies. Despite being founded in 1946, their eyes are fixed firmly on the future as they undergo a major digital transformation drive, and it is here that Zindi comes in.

“It’s clear to us that transformation on the digital side will be a key enabler of our business, and the level of talent that we can get through Zindi is unmatched,” says Mustafa Sachak, Group Chief Executive Officer for Zimnat. “Digitilisation has accelerated tremendously during the pandemic, and we need to make sure that we on the continent are following the same path.”

He also emphasises the value of partnerships (one of Zimnat’s core values) in this space.

“Zindi has given us access to data scientist talent across the continent — something we would never be able to achieve as one company. We’ve had some great solutions come out of our competitions with Zindi — solutions for churn prediction, a recommendation engine, and estimating loan defaults for our microfinance business.”

Growing partnership in data science

While the story of Zimnat and Zindi started with a #ZindiWeekendz weekend hackathon, it has blossomed into much more. Zindi and Zimnat have run a further two competitions since — the Zimnat Insurance Recommendation Challenge (a three-month challenge with a US$5000 prize pot), and UmojaHack Zimababwe: Zimnat Loan Default Prediction (a machine learning hackathon for Zimbabwean university students).

With similarly rapid turnaround times, winning models for both challenges are already being put into production by Zimnat’s digital team, and Zulu hopes that these products will be live in the first quarter of 2021. He is particularly excited about the loan default prediction model.

“Most of our customers don’t have a formal credit rating,” he explains. “The models we received from Zindi allow us to look at certain aspects of the customer based on the data we have, and say ‘this person qualifies for this amount of credit’. It will change how we deliver our microfinance products.”

For us at Zindi, seeing winning machine learning models from all over Africa being implemented so quickly, with a directly impact on business outcomes for our clients, gets us excited and makes our hard work worthwhile. We’re proud of our partnership with Zimnat, and look forward to a long and fruitful relationship together.

Zindi is a competition platform hosting a community of data scientists dedicated to solving Africa’s toughest challenges through machine learning & AI

Get the Medium app

A button that says 'Download on the App Store', and if clicked it will lead you to the iOS App store
A button that says 'Get it on, Google Play', and if clicked it will lead you to the Google Play store