The Challenge
RD Station Marketing is one of Brazil's leading marketing automation SaaS platforms. The product team had an ambitious backlog of 30+ growth experiments focused on retention, engagement, and conversion, but they wanted results within weeks, not months. The challenge was prioritizing the highest-impact experiments, executing them quickly in a React and Ruby on Rails stack, and measuring results rigorously.
My Role
As Senior Frontend Engineer, I was responsible for:
- Frontend implementation of growth experiments across the platform
- Collaborating with product and growth teams to prioritize experiments by impact
- Building and deploying A/B tests with proper measurement
- Redesigning the pricing page to optimize plan conversion and upsell
- Rebuilding the onboarding flow to boost paid user engagement
Technical Approach
Prioritization Framework
With 30+ experiments on the backlog, I worked with the product manager and growth team to stack-rank experiments using an impact-vs-effort framework. We identified the top three highest-impact opportunities:
- Pricing page redesign - Direct impact on plan conversion and upsells
- Onboarding flow rebuild - Impact on paid user engagement and retention
- Feature discovery experiments - Driving users to underutilized features
Pricing Page Experiment
The pricing page was the highest-leverage point in the conversion funnel. I redesigned it with:
- Clear value differentiation between plans
- Social proof and trust signals to reduce friction
- Streamlined upgrade flow to capture upsell opportunities
- A/B tested variations to validate each change with data
Onboarding Flow
The existing onboarding was losing paid users before they experienced the product's value. I rebuilt it with React and Ruby:
- Progressive disclosure - Guided users to key features step by step
- Personalized paths - Different flows based on user goals
- Engagement triggers - Strategic nudges to drive meaningful actions
Experiment Velocity
To ship 30+ experiments efficiently, I established:
- Reusable experiment components - Common patterns for A/B tests
- Feature flags - Quick toggling without redeployment
- Measurement templates - Consistent tracking across experiments
- Weekly prioritization reviews - Adjusting based on learnings
Key Learnings
- Focus on impact, not volume - The top 3 experiments drove more value than the other 27 combined
- Data beats opinions - A/B testing resolved internal debates quickly
- When workload feels impossible, prioritize ruthlessly - Saying no to low-impact work protected quality
Results
- 7% increase in plan conversion - from ~2.90% to ~3.19%
- 12% increase in upsell rate - from ~2% to ~2.24%
- 66% boost in monthly engagement - from 150 to 250 actions per paid user
- 30+ experiments deployed with proper measurement and iteration