Keeping students learning in the age of AI through behavioral segmentation

AI is changing how students study faster than learning platforms are changing how they serve students, and tools that don't adapt lose them. A leading online learning platform asked us to map how students actually use online educational tools, what drives them to subscribe or leave, and where the unmet needs are, to steer product strategy and positioning.

We interviewed students from high school through graduate school, using thematic analysis to surface the drivers, barriers, and gaps in how they use learning tools, then built the findings into a survey of more than 800 students. Unsupervised machine learning revealed that willingness to subscribe, an attitude, not a demographic, produced the cleanest and most predictive segmentation, yielding six distinct segments: from committed subscribers to unengaged non-users, plus three distinct clusters within high school students.

We paired the analysis with an interactive dashboard so the client's teams could explore the data themselves, then ran two concept development workshops with the client to turn the highest-value opportunities into product decisions.

Six behavioral segments now describe the platform's users and non-users, each with its own barriers, drivers, and unmet needs. The workshops produced two fully developed product concepts, complete with strategic principles, key attributes, KPIs, and risk assessments, plus ten further product ideas, and the client keeps the dashboard as a living tool for exploring its own market.