Reducing churn for a European media company
.svg)
Reducing churn by 30% in 3 months: how a leading media company used AI-powered analytics to reveal thousands of clues that identify which reader was likely to churn, and why.
The Challenge
Sweden’s leading media company’s digital subscription base was stagnating: 14.8% monthly churn nearly overcame its 10-15% monthly growth.
In their words, they were 'busy dying'. The Chief Editor reached out to SparkBeyond to understand the root causes of the problem, and build a sustainable business model with targeted proactive retention activity.
SparkBeyond helped infuse Dagens Nyheter with artificial intelligence (AI) and methodologies that identified patterns — to a level whereby we can identify within 86% accuracy who will likely churn.
The Approach
SparkBeyond Discovery was embedded into a cross-functional team of business stakeholders and data analysts. The team used the Discovery platform to automatically augment diverse internal datasets (subscription, billing, status) with external data sources (Wikipedia, census).
Within a week, the team identified >200 actionable drivers of churn, which were then featured on a real-time dashboard, empowering the media company with fast iterations to take action every day.
Using data-driven mapping to quantify and select high-potential initiatives, the cross-functional team built a model that accurately predicted 86% of churners.
Drivers of high churn:
- Subscribers receiving monthly reminders (paper/digital invoice)
- Subscribers acquired via telemarketing
Drivers of low churn:
- iPhone & iPad users
- Subscribers for longer periods (i.e.students)
- Subscribers with electronic and automated payments
Results
By testing millions of patterns per minute, SparkBeyond Discovery revealed the root causes behind the cohort of customers who churn.
Examples include:
- Cross-page links in the iPhone app crashed -- leading to disengagement and churn.
- Monthly invoice reminders led to churn
Peter Wolodarski and his team were able to interpret the root causes and take action with the following initiatives:
- Create autopay flow
- Migrate stock to autopay
- New sales channels
- Telemarketing optimization
- Broader payment options
- Subscription optimization
By employing the above actions, the media company reduced its churn by 30% within a week.
As new root causes emerged over time, the team ran a hackathon every two weeks to interpret emerging patterns and continuously improve its retention activity. This once-struggling media company wasn't busy dying: it was busy growing.
Features
Related Case Studies
It was easier in this project since we used this outpout
Business Insights
Apply key dataset transformations through no/low-code workflows to clean, prep, and scope your datasets as needed for analysis
Predictive Models
Apply key dataset transformations through no/low-code workflows to clean, prep, and scope your datasets as needed for analysis
Micro-Segments
Apply key dataset transformations through no/low-code workflows to clean, prep, and scope your datasets as needed for analysis
Features For
External Models
Apply key dataset transformations through no/low-code workflows to clean, prep, and scope your datasets as needed for analysis
Business Automation
Rules
Apply key dataset transformations through no/low-code workflows to clean, prep, and scope your datasets as needed for analysis
Root-Cause
Analysis
Apply key dataset transformations through no/low-code workflows to clean, prep, and scope your datasets as needed for analysis
