Product thinking / Case study

Conversion Funnel Optimisation - Growth

Company: AECC Global | Role: PM - AI & Growth | Year: 2024

Owned the end-to-end funnel from student inquiry to completed application. Improved meeting-to-applicant conversion from 15% to 18% YoY - representing ~3,000 additional applicants annually.


The Problem

AECC’s core revenue model is conversion: an inquiry becomes a meeting, a meeting becomes an application, an application becomes a commencement (enrolled student). The biggest drop-off in the funnel was meeting → applicant. At 15% conversion:

  • For every 100 students who had a counsellor meeting, 85 did not submit an application
  • The cost of acquiring that meeting (marketing spend, counsellor time) was not being recovered
  • Small conversion improvements had outsized impact on revenue

Discovery: Understanding the Drop-off

Funnel teardown

I built a full funnel map in the CRM (Zoho) with drop-off rates at each micro-step:

Meeting scheduled → Meeting attended → Follow-up sent → Documents requested → Application started → Application submitted

Key finding: the biggest drop was after “meeting attended” - 60% of drop-off happened in the first 48 hours post-meeting.

Interview findings (30 student interviews)

Top reasons for not applying after a meeting:

  1. “I wasn’t sure which course was right for me” (confusion / analysis paralysis)
  2. “The counsellor didn’t follow up quickly enough” (responsiveness)
  3. “The application process felt overwhelming” (complexity friction)
  4. “I’m still comparing options” (decision stage mismatch)

Counsellor interviews (20)

  • Average follow-up time post-meeting: 28 hours (target was same-day)
  • Counsellors had no structured post-meeting checklist - every counsellor had a different follow-up flow
  • No CRM automation to prompt follow-up or escalate cold leads, only static dashboards

Solutions Designed & Shipped

Initiative 1: Post-Meeting Follow-Up Automation

The insight: 60% of drop-off was in 48 hours. Speed of follow-up was the #1 controllable variable. What we built:

  • Automated WhatsApp + email sequence triggered 2 hours after meeting ends
  • Personalised with counsellor name, course shortlist summary, and direct application link
  • Counsellor CRM task auto-created for same-day follow-up call
  • Escalation to team lead if no counsellor action within 24 hours Result: Average follow-up time: 28 hours → 6 hours

Initiative 2: Post-Meeting Summary (Meeting Insights Generator)

Separate case study - but the Meeting Insights Generator was partly motivated by this funnel work. Students who received a personalised summary of their meeting converted at 2.1× the rate of those who didn’t.

Initiative 3: Application Kickstart Flow

The insight: “The application process felt overwhelming” = friction, not lack of intent. What we built:

  • Simplified onboarding flow for students who clicked the application link
  • Progress indicator showing “3 steps to complete” instead of showing the full form upfront
  • Document pre-check (via AI Document Intelligence) to reduce back-and-forth
  • WhatsApp nudge if application started but not submitted after 24 hours

Initiative 4: Counsellor Workflow Redesign

The insight: Inconsistent counsellor quality was a major variance driver. What we built:

  • Standardised post-meeting checklist in CRM (checklist-style completion, not free text)
  • Course shortlist auto-generated and attached to follow-up (from Smart Recommender)
  • Performance dashboard per counsellor - visible to counsellor and their manage.


What I Learned

  • Funnel work is systems thinking - the drop-off wasn’t one problem, it was 4 interconnected problems requiring coordinated fixes
  • Speed > perfection in follow-up - a mediocre follow-up in 2 hours outperformed an excellent follow-up in 2 days
  • Counsellor behaviour change required incentives, not just tools - adding the performance dashboard (with manager visibility) drove more behaviour change than the new checklist alone
Explore more work