How Mega Energy achieved 22% reduction in customer churn and increased LTV with Churned & Eneve's GridHub
The challenge
Mega Energy is one of the fastest-growing energy suppliers in the Netherlands. Like many players in the industry, they are rapidly acquiring customers, through partnerships, aggressive offers, and switching platforms.
But growth also brings churn.
Retaining customers beyond the first year, and increasing their lifetime value, has become one of Mega’s top commercial priorities. They didn’t just want insights, they needed a way to act on them at scale. And they wanted to do that without slowing down internal teams or launching complex new tooling on top of their existing systems.
Churned worked closely with the Mega team, including Ninjiro Helstone (Team Lead Operations), Noah Barendse (Manager BTC), and commercial leadership, to make that happen.
Why Eneve's GridHub?
GridHub is a SaaS platform and one of many Eneve solutions tailored for the energy market. Based in Utrecht, it supports over 18 energy suppliers with contract processing, customer lifecycle tooling, and acquisition flows.
For Mega, GridHub is the operational core. Everything runs through it, from CRM and ERP data to customer communications.
That’s why it made sense to connect the retention solution to GridHub.
Why Churned?
Churned is directly integrated into GridHub. It connects to the customer data stored in GridHub and starts generating predictions for churn, upsell, and reactivation.
But prediction is only the starting point.
Churned also identifies which action works best for each profile, based on A/B test results and historical outcomes, and then automates those actions across the customer base.
This was key for Mega: not just dashboards they had to interpret, but added intelligence, models that learn and act.
"We wanted to move beyond insights and dashboards, to actually do something with the data."
Noah Barendse, Mega Energie Netherlands
The solution in action
Mega connects its customer and contract data from GridHub directly to Churned. Churned’s AI models run over that data to predict which customers are at risk of churning, and more importantly, which action is likely to retain them, whether that’s a discount, contract renewal offer, or simply leaving them untouched.
But the model doesn’t stop at recommendations. It tests them.
Each next-best-action is tested against a control group using A/B methodology. For every campaign, the system compares the outcomes between customers who received Churned’s recommended action and those who followed the old strategy.
The model learns from those experiments, identifying not just which action works best overall, but which actions work best for specific customer profiles, and when to apply them.
This approach ensures that every decision is backed by actual retention impact, not guesswork. The model validates its own performance, proving whether its next-best-actions actually outperform the status quo before scaling them.
"The model doesn’t just tell us who’s at risk, but gives us something we can actually execute and trust."
Ninjiro Helstone, Mega Energie Netherlands
Results
The A/B test setup allowed for a clean comparison between the AI-driven group and the control group. Same moment, same message, same customer type.
- Churn rate (AI group): 3.47%
- Churn rate (control group): 4.48%
→ That’s a 5% reduction in churn compared to the control group. - Conversion rate (AI group): 2.7%
- Conversion rate (control group): 2.4%
→ A 5% increase in conversion.
These are measurable lifts. Two additional campaign types now have enough data to deploy the same model-based automation.
What's next?
Possibly scaling the approach to more segments and campaigns within GridHub.
By combining GridHub’s infrastructure and Churned’s intelligence, they’ve created a setup where experimentation, learning, and execution all happen in one flow.
