Introduction
Scaler's Data Science and Machine Learning program is a long online program for working professionals, combining live classes, weekly assignments, projects and mentorship. Assignment completion is the best early predictor of whether a learner finishes the program and achieves a career outcome. The scenario: in the latest cohort, weekly assignment completion in month three fell from a historical 72% to 58%. Earlier months looked normal. Leadership wants to know why before the next cohort reaches the same stage. This is a hard root cause question because the cause could sit in curriculum, learner life events, product changes, or the shift towards AI content across Scaler's programs.
In learning products, separate "can't" (too hard, unclear), "won't" (motivation) and "didn't know" (product or communication issues). Each needs a different fix.
Step 1
Clarifying Questions (3 minutes)
- Hypothetical answer: Share of active learners who submit the weekly assignment with a passing score before the deadline.
- Impact: A change in difficulty, deadlines or scoring rules could move it.
- Hypothetical answer: Starts in week 9, when the cohort enters the statistics and machine learning foundations module, and persists.
- Impact: Points to the module.
- Hypothetical answer: Yes. This cohort's curriculum was refreshed: some statistics content was compressed to make room for a new generative AI module, and assignments now use a new auto-graded notebook environment.
- Impact: Two strong candidate causes.
- Hypothetical answer: More learners from non-tech backgrounds joined this cohort after a marketing campaign targeting career switchers.
- Impact: Audience mix could change completion.
- Hypothetical answer: Live class attendance fell slightly (−5%); doubt-session requests rose 40%.
- Impact: Learners are engaged but struggling, more "can't" than "won't".
Practice similar questions
Subscribe to access the full answer
.png){kind=logo}
