Benefits & Results

Significant drop in churn rates as predicted customers were engaged proactively
Enhaanced communication strategies led to better customer satisfaction and loyalty
More stable customer base reduced pressure on acquisition teams, leading to optimized acquisition spends
Improved retention directly contributed to higher and more stable revenue

Achieved a high retention rate of 11% over a period of three months based on churn identification and retention strategy defined by marketing teams, resulting in revenue 'savings' of £700,000

Background

Nuffield Health is a leading healthcare and wellness brand in the UK, running multiple facilities including private hospitals, fitness, and wellbeing centers. They envision providing a 'Connected health and wellbeing service' by offering an end-to-end range of health experts and services to cover all customers' health needs - physical, nutritional, or emotional, ensuring accessible and personalized care by their clinicians either face-to-face or through digital channels.

‍

Challenges

Issue Identification: Nuffield Health's gyms lacked a comprehensive understanding of the underlying reasons for their high churn rate and the inability to identify potential churners to apply relevant marketing interventions

Issue Impact: High churn led to an annual revenue loss of £6 million and high investment in acquisitions due to lower customer retention, leading to overall high operational costs. This adversely affected long-term planning and resource allocation as the organization struggled to maintain a consistent customer base and annual revenue.

Solution

NowVertical developed a comprehensive churn prediction solution leveraging advanced data science techniques and predictive analytics to identify at-risk customers. Dashboard visualizations generated actionable insights for Retention and Marketing teams to take pre-emptive churn mitigation interventions to avoid gym membership cancellations.

Implementation

Data Collection and Integration: Gathered data from various sources including customer interaction logs, membership details, and service usage patterns.
Data Analysis: Utilized Python and AWS (Postgres) to process and analyze the data, identifying key indicators of churn
Model Development: Developed a churn prediction model using machine learning algorithms validated against historical data
Dashboards: Developed visualizations and reports to identify 'churn risky' gym members of each customer cohort
Actionable Insights: Generated actionable insights and recommended interventions to reduce churn such as personalized retention communications and personalized offer recommendations