Shortly before I joined, Intruder had just launched a new onboarding flow, it was rigid and assumption-based, with no post-release validation.

Only 3% of users converted, with most dropping off before step two. Success was measured only by top-line outcomes, providing no insight into user behaviour or friction points.

We began with a analytics deep-dive that revealed a 96% drop-off rate during onboarding. Session recordings added to this by showing that users wanted to skip ahead and explore directly, rather than being forced through each step.

Competitor analysis confirmed our suspicions, their flows were far shorter and more flexible, while ours dragged on with 21 rigid steps. Finally, both customer interviews and stakeholder feedback reinforced the same message: the onboarding experience was obstructive.

To validate our findings, we designed an A/B test.

• Control group experienced the existing 21-step onboarding flow, the same rigid journey we knew was causing high drop-off.
• Test group had onboarding disabled entirely. Instead, they were dropped directly into the product’s empty state, which simply prompted them toward the first key action: Add a target.

This test isolated the effect of removing friction altogether, allowing us to compare whether users made faster progress—and with less drop-off—when guided by a minimal entry point rather than a lengthy prescribed flow.

Impact

30% immediate increase in conversion, followed by iterative improvements that ultimately increased conversion from 3% to 9.5% - a 177% improvement.

This validated that reducing friction, not adding structure, drove better outcomes. We immediately launched a lightweight redesign improving the empty state experience.

Key Design Shifts

We moved away from linear flows and toward contextual suggestions, so users get help only when they need it, not all at once upfront.

Instead of driving users toward step completion, we reframed onboarding around meaningful first actions, prioritising value delivery over box-ticking.

And where the old experience created immediate friction, we shifted to deferred complexity, introducing advanced features only after users had grasped the core value of the product.

Our research revealed that time to first report strongly predicted engagement, leading to the proposition of additional tests to capitalise on the gains here:

• Auto-adding user email domains as scan targets
• Introducing faster scan types for quicker results
• Previewing example reports during initial scans
• Increasing trial access to additional (premium) scanners

The complete redesign ultimately improved conversion from 3.25% to 9% - a 177% increase.

By June 2025, we achieved significant year-over-year gains across the onboarding funnel through cultural shift toward experimentation and evidence-based iteration.

Post-release evaluation became standard practice, establishing a foundation for scalable improvements and data-informed decision making.

One of the biggest lessons was that less can be more, the strongest design decisions often come from knowing what to remove, not just what to add.

We also saw the power of putting data over opinions. User insights and metrics consistently led to better decisions than assumptions or internal debate.

Another takeaway was the importance of value over process. Onboarding should shorten the path to core value, rather than serving as a showcase of every feature.

Finally, we learned that culture change doesn’t always come from big launches. Even lightweight experiments can shift organisational mindset and open the door to new ways of thinking.