Building a nurture flow for SaaS onboarding

Context

As both the Director of Marketing and (an aspiring) developer at Capptions, I noticed a gap in our user onboarding process. New users were receiving the same generic welcome emails regardless of their behavior or engagement level. This one-size-fits-all approach wasn't effectively guiding users through our platform's key features, particularly our marketplace and DIY workflow builder.

The challenge was clear: build an intelligent nurture flow that could adapt to user behavior, provide relevant guidance, and ultimately drive better activation rates. I wanted to create something that felt personal and helpful, not just another automated email sequence.

My constraints were:

What I Built

I developed a dynamic email nurture flow in HubSpot that:

  1. Adapts to user behavior with 5-7 personalized touchpoints
  2. Segments users based on their engagement patterns (marketplace buyers vs. DIY users)
  3. Tracks specific in-app actions (credit usage, workflow completion, form/report creation)
  4. Triggers contextual help emails based on user progress
  5. Automates CRM updates and internal notifications

The system uses HubSpot's workflow capabilities with our web app's integration to create a responsive onboarding experience.

Nurture flow minimap

Technical Breakdown

Stack & Tools

Key Architecture Decisions

  1. Use-case Based Branching Logic
Use-case based branching logic
  1. Usage Based Branching Logic
Usage based branching logic

Edge Cases Handled

  1. Multiple User Paths

    • Marketplace vs. DIY user detection
    • Credit usage tracking
    • Workflow completion status
  2. Email Frequency Control

    • Minimum 24-hour gaps between emails
    • Maximum 7 emails per user
    • Disallowed enrolment of users (domain-based) upon CSM request

What I Learned

  1. Email Performance Varies by Context

    Our initial welcome email metrics showed:

    • Open Rate: 25.98%
    • Click Rate: 4.8%
    • Unsubscribe Rate: 0.53%

    This baseline helped us understand where to focus our optimization efforts.

  2. Content Type Significantly Impacts Engagement

    Different types of content showed varying levels of engagement:

    Product Education Emails:

    • "Understanding your Dashboards": 29.48% open rate
    • "How to use purchased bundles": 50% open rate

    Feature Announcement Emails:

    • "Marketplace templates": 30.67% open rate
    • "Digitize your form": 16.06% open rate

    The data clearly showed that practical, how-to content consistently outperformed feature announcements.

  3. Key Optimization Learnings

    Through A/B testing and iteration, we identified several patterns:

    a) Timing Matters

    • Emails sent after specific user actions (like bundle purchases) saw higher engagement
    • "How to use purchased bundles" achieved 50% open rate and 8.82% click rate
    • Immediate post-purchase timing proved more effective than delayed sends

    b) Subject Line Impact

    • Direct, action-oriented subjects performed better
    • Including the word "Dashboard" improved open rates by ~10% (29.48% vs baseline)
    • Variations of "How to create an Organization" showed consistent ~25% open rates

    c) Content Strategy

    • Task completion emails ("Complete your form") saw highest open rates (55.17%)
    • Technical guidance emails maintained steady ~18% engagement rates
    • Product education content needed to be highly specific to maintain engagement
  4. Unsubscribe Rate Patterns

    We tracked unsubscribe rates carefully:

    • Early emails: 0.53% unsubscribe rate
    • Mid-flow emails: 1.5-3% unsubscribe rate
    • Later emails: 3-10% unsubscribe rate

    This led to two key improvements:

    • Implemented better user segmentation to reduce irrelevant sends
    • Added clearer expectations about email frequency in welcome message
  5. Skip Rate Insights

    The skip rate data proved particularly valuable:

    • Early emails: < 1% skip rate
    • Technical guides: 2-3% skip rate
    • Later workflow emails: 13-18% skip rate

    This helped us:

    • Optimize email sequence ordering
    • Identify when users were ready to graduate from the nurture flow
    • Better time our CSM intervention points

Implementation Improvements

Based on these learnings, I've made several tactical changes:

  1. Content

    • Shortened email copy by 30%
    • Added clear, single-action CTAs
    • Included specific use-case examples
  2. Timing

    • Reduced frequency for users with high engagement
    • Added action-based triggers for key feature announcements
    • Implemented smart delays based on user timezone
  3. Segmentation

    • Created separate flows for marketplace vs. DIY users
    • Adjusted content based on user's interaction history
    • Implemented engagement-based branching logic

What's (possibly 🤪) Next?

  1. Personalization

    • Industry-specific content paths
    • Role-based recommendations

    (both only possible upon bettering our in-app onboarding + direct integration)

  2. Content

    • Systematic subject line testing (e.g. "How to create an Organization" vs "Creating an Organization is easy!")
    • Send time optimization (delay vs trigger based)
    • Content format experimentation (e.g. images, videos, etc.)
  3. Analytics

    • Deeper cohort analysis (limited by Hubspot's capabilities)
    • Engagement patterns analysis
    • Churn prediction integration (may be overkill for now as we're still learning)

The most valuable insight from this project is that email engagement isn't just: open rates / click rates in isolation - it's about sending the right content to the right user at the right time so that the user derives value from the product. Our most successful emails weren't necessarily those with the highest open rates, but those that led to meaningful product engagement and user activation. 🧀


Note: All metrics are from actual production data from Q1 2025. While some metrics may seem modest compared to industry benchmarks, they represent real-world performance in the B2B SaaS space where engagement patterns differ significantly from B2C or general marketing emails.