Selected projects & case studies
Most of my work is behind the scenes, but here are a few representative examples. Details are anonymised where needed.
Bringing clarity to an admissions funnel
Private University • 6 months engagement
The problem
A private university had thousands of leads coming in from multiple sources – website forms, calling campaigns, walk-ins, and referrals. But they had no clear visibility into:
- Which sources were bringing quality leads
- Where leads were getting stuck in the funnel
- How individual counsellors were performing
- Which campaigns were actually worth the investment
What I did
I built a full admissions analytics layer that connected their CRM, calling logs, and admissions data into a single coherent view.
Key components:
- Lead source tracking and attribution model
- Funnel stage definitions and conversion tracking
- Counsellor performance dashboard with calling metrics
- Campaign-level ROI analysis
- Lead quality scoring based on historical patterns
- Weekly automated reports for leadership
The outcome
The admissions team finally had a clear picture of what was working. They could see exactly where leads were dropping off, which counsellors needed support, and which marketing channels deserved more investment. Decision-making went from gut-feel to data-backed.
Tools used: Excel, Power BI, SQL, Python for data processing
Recruitment manager for online programs
EdTech Startup • 3 months project
The problem
An online learning platform was scaling their recruitment team and losing track of candidates. They needed:
- A simple way to track candidate status and interview stages
- Automated follow-up reminders for recruiters
- Meeting scheduling and calendar integration
- Manager visibility into team activity
What I did
I designed and built a recruitment manager using Google Apps Script and Sheets – keeping it simple and familiar for the team.
Key features:
- Candidate tracking with customizable stages
- Automated email reminders for follow-ups
- Google Calendar integration for interviews
- Dashboard for managers showing pipeline health
- One-click status updates and notes logging
- Weekly summary reports via email
The outcome
The recruitment team went from scattered spreadsheets and missed follow-ups to a centralized system everyone could understand and use. Managers had real-time visibility, and candidates had a better experience with timely communication.
Tools used: Google Sheets, Google Apps Script, Gmail API, Google Calendar API
Lead quality & calling policy insights
Admissions Consulting Firm • 2 months analysis
The problem
An admissions consulting firm was making thousands of calls but didn't know:
- How many attempts were optimal before giving up
- What time of day had the best connection rates
- Which lead characteristics predicted conversion
- Whether their calling policy was actually effective
What I did
I analysed 6 months of calling data and outcomes to find patterns and recommend policy changes.
Key analyses:
- Call attempt analysis – optimal number before drop-off
- Time-of-day and day-of-week connection patterns
- Lead quality indicators from source and demographics
- Counsellor script effectiveness comparison
- Follow-up timing impact on conversion
The outcome
The firm updated their calling policy based on clear evidence. They reduced wasted effort on low-quality leads, optimized calling schedules, and improved overall conversion rates by focusing on what actually worked.
Tools used: Python (pandas, matplotlib), Excel for reporting
CRM cleanup & workflow redesign
Professional Services Firm • 4 months engagement
The problem
A professional services firm had been using their CRM for 3 years, but it had become a mess:
- Duplicate records everywhere
- Inconsistent status labels and stage definitions
- No clear ownership or assignment rules
- Reports that nobody trusted
What I did
I led a full CRM cleanup and workflow redesign to get them back on track.
What we accomplished:
- Cleaned and deduplicated 15,000+ contact records
- Redesigned stage definitions with clear criteria
- Created lead assignment and routing rules
- Built new dashboards based on clean data
- Trained the team on new workflows and best practices
- Documented everything for future reference
The outcome
The team finally had a CRM they could trust. Reports were accurate, processes were clear, and everyone knew what they were supposed to do. The system went from being an obstacle to being a real asset.
Tools used: CRM platform (various), Python for data cleaning, Excel for validation
Automated call notes analysis
University Admissions Team • Proof of concept
The problem
A university admissions team was making hundreds of calls every day. Counsellors were logging notes, but nobody had time to read through them all. Important insights and patterns were buried in text.
What I did
I built a proof-of-concept AI system that automatically analysed call notes and extracted key information.
Features:
- Automatic categorization of call outcomes
- Extraction of common objections and concerns
- Identification of urgent follow-ups
- Sentiment analysis of student interactions
- Daily summary reports for managers
The outcome
Managers could finally see patterns across hundreds of calls without reading every note manually. They spotted issues faster, identified training needs, and improved response times for urgent cases.
Tools used: Python, OpenAI API, custom scripts
Want to discuss a project?
If you have a similar challenge or need something different, let's talk about what might work for your context.