Short-Video Funnel Diagnostic
Views but no sales? Enter your numbers and find out which stage of the journey is losing customers: reach, content, credibility or purchase, and which stages cannot be judged yet because there is too little data.
Results
Stage Events / trials Rate 95% interval Normal rate Verdict
How it works. Stages are judged in order, from reach to purchase. Each conversion rate gets a 95% Wilson confidence interval. A stage is flagged when your normal rate lies outside that interval. A stage with fewer than about 60 successful transitions is marked not yet diagnosable unless the gap is already clear: a rule of thumb that keeps the relative error near ±25%, not a universal threshold. Your normal rate is itself an estimate, so treat close calls with care.
This is a diagnostic aid, not a causal test. Confirm any diagnosis with a controlled comparison, such as matched video variants.
Method: Emelianova, M. (2026). Diagnosing conversion failure in short-video commerce: A stage-based framework for localising drop-off across reach, content, credibility and purchase. Zenodo. doi.org/10.5281/zenodo.23142065
All calculations run in your browser. No data is sent anywhere.