Centers of Enablers in Experimentation with Ruben de Boer

Centralized Model

In the realm of Conversion Rate Optimization (CRO), the centralized model stands out for its strong knowledge-sharing capabilities. Ruben de Boer explains that this model involves a single team handling all experimentation, ensuring clear responsibility for tools and recruitment 1. However, a significant drawback is the team's distance from product teams, which can hinder innovative experiments. This separation often limits the team to small optimization tasks, as they lack specialization in specific product areas 1. Ruben highlights that while this model fosters a cohesive team, it may not always align with the dynamic needs of product development.

   

Hybrid Model

The hybrid model, or center of excellence, merges centralized and decentralized elements to enhance experimentation success. Ruben advocates for this approach, where CRO specialists are embedded within product teams, becoming domain experts while supported by a central team 1. This setup facilitates both large and small experiments, ensuring quality and accessibility across the organization. Ruben prefers the term "center of enablers," emphasizing the role of empowering others to conduct experiments 2. He notes, "You're not the one presenting the brilliant A/B test to your colleagues anymore. You enable others to run A/B tests." This model is particularly effective in fostering a culture of experimentation.

   

Adapting Models

Adapting experimentation models to fit organizational size is crucial for effective CRO strategies. Ruben suggests that smaller teams might benefit from a centralized approach, gradually evolving into a center of excellence as they grow 3. This step-by-step maturity process helps organizations develop their experimentation culture and strategy. Education plays a pivotal role in this transition, as tailored training enhances the ability to run high-quality experiments 4. Ruben emphasizes the importance of proactive education, stating, "Without knowledge, it's very hard to run experiments." This adaptability ensures that organizations of all sizes can cultivate a robust experimentation framework.