Work
Each project is written up as a research case study: the question, the data, the method, the honest limitations — and what I'd do differently next time.
Online learning platforms lose students silently — by the time dropout is visible, it is often too late to intervene.
“Behavioural signals in activity logs can be turned into an early-warning perspective on dropout — before the student is already gone.”
Read case study
A data-rich platform needs to work for analysts and non-technical decision makers alike — and be findable by both search engines and AI assistants.
“Analytics shows what users do; research explains why. Recommendations only become convincing when both line up.”
Read case study
An ed-tech company had rich usage data from its learning platform, but no comparable measure of how difficult its subjects actually were for learners.
“A good index compresses messy behavioural data into one number the client can actually reason with — without hiding what went into it.”
Read case study
Social content decisions are usually made on instinct — this project asked what the engagement data actually supports.
“Engagement metrics only become insight when connected to content characteristics and audience behaviour, not counted in isolation.”
Read case studyShowing 4 of 4 projects
Contact
I’m open to early-career opportunities in data science, analytics, UX research, and insights — and always happy to talk about interesting problems.