Back to Projects

Personalizing Thrive’s Onboarding to Drive 23% More First-Week Engagement

Streamlined a noisy, one-size-fits-all onboarding flow into a focused, personalized experience that increased early engagement and reduced support friction.

Onboarding Personalization Enterprise SaaS Scale
23%More first-week coaching actions
12%Decrease in onboarding drop-off
11%Decrease in support tickets

Context

Why are we here?

Thrive Global is a B2B SaaS platform designed to help organizations support employee well-being through personalized behavior-change programs. As the enterprise product matured, we noticed early onboarding was doing too much: it front-loaded information and asked users to understand the system before they’d experienced value.

Thrive Global hero image
Example: existing UI surfaces multiple paths at once, increasing cognitive load.

Problem

What are we trying to solve?

The onboarding flow was overloaded with information and lacked personalization, causing users to:

  • Struggle to tailor Thrive Global to their personal needs (too many options presented too early).
  • Take fewer meaningful actions, leading to minimal engagement with the core tool.
  • Feel overwhelmed after onboarding with unclear guidance on what to do first.

This led to slower habit-building and made it harder for users to discover the full value of Thrive.

Evidence

How do we know it’s a problem?

Through a combination of product analytics, user research, and customer feedback, we uncovered:

  • Drop-off during onboarding before users reached their first “value moment.”
  • Session recordings showing confusion and decision fatigue as users navigated too many choices.
  • User interviews revealing employees wanted to set clear boundaries (e.g., reminders/check-ins) and focus on a single priority first.

We also heard from HR leaders that onboarding often felt like “another cognitive load,” not a supportive experience.

My Role

As the lead designer for Thrive’s onboarding initiative, I focused on building a scalable foundation for structured collaboration across teams, improving efficiency in delivery, and delivering a best-in-class onboarding experience for enterprise clients.

Key responsibilities included:

  • 🔄 Ran end-to-end design - from discovery and ideation through high-fidelity design and developer handoff.
  • 🧠 Facilitated workshops - with PM, engineering, and stakeholders to define key moments of personalization.
  • 🧱 Scaling with the design system - creating modular onboarding components that could be adapted across enterprise setups.
  • 🧪 Conducted moderated usability tests - validating flows, messaging, and recommendation patterns.
  • 🤝 Partner with Customer Success - ensuring the experience aligned with enterprise rollout needs and communication styles.

Team

1 PD (me), 1 PM, 1 Engineer, 1 Data Analyst, Customer Success partners

Timeline

~5 months (discovery → shipped experiments → iteration)

Solution

What’s the suggested solution?

We streamlined and personalized the onboarding experience by introducing:

  • A “personal selector” upfront so users can choose a primary intent (focus), reducing noise and speeding up time-to-value.
  • Preferences & reminders that let users set boundaries (when/how they want to be nudged).
  • A tailored starting dashboard that reflects chosen priorities immediately after onboarding.
Onboarding intent selector
Intent selection: reduce decision fatigue by asking one clear question first.
Preferences and reminders
Preference controls: empower users to set boundaries and opt into reminders.
Preferences and reminders
Progressive disclosure: breaking onboarding into smaller, context-aware moments instead of overwhelming users with everything up front.
Dynamic dashboard
Personalized dashboard: the seeds that reflected chosen priorities right after onboarding, giving an immediate sense of relevance.
Design principle

Ask one meaningful question early, then progressively disclose options as the user gains confidence.

Impact

Quantitative

  • 12% decrease in onboarding drop-off rates, improving overall activation.
  • 23% more users completed a first coaching activity within 7 days, indicating higher early engagement.
  • Reduced support tickets related to onboarding experience by 11%.

Qualitative

  • User feedback highlighted the value of customization.
  • Customers reported onboarding felt “less noisy” and helped employees focus on what mattered first.
  • HR partners felt more confident rolling out Thrive to new hires due to clearer guidance and a more supportive start.

Belief

Why do we believe it will work?

Behavior-change products work best when they reduce friction and make the “first right step” obvious. By guiding users toward a single priority, letting them set boundaries, and showing immediate relevance, we improved clarity without removing choice, we simply sequenced it.

New onboarding flow
New onboarding flow.
Old onboarding flow
Old onboarding flow.

Let’s build something great.

Always open to connecting about roles, systems work, and solving messy problems.