Binge-Watching Pattern Mining | Dataford Interview Questions

Netflix wants to identify recurring binge-watching sequences from profile viewing logs. A binge pair is two completed sessions for the same profile where the second title starts after the first session ends and within 30 minutes.

Write a PostgreSQL query to find and rank title-to-title binge patterns.

Requirements

Self-join viewing_sessions to compare consecutive sessions for the same profile.
Include only completed sessions whose start times meet the 30-minute rule.
Keep patterns observed at least twice, and calculate pair count and total minutes watched.
Rank each profile's patterns by pair count descending, then total minutes descending. Return the results in profile and rank order.

Schema

viewing_sessions

Column	Type	Description
session_idPK	INT	Unique viewing session identifier
profile_id	INT	Netflix profile identifier
title_id	INT	Watched title identifier
started_at	TIMESTAMP	Session start timestamp
ended_at	TIMESTAMP	Session end timestamp
duration_minutes	INT	Minutes watched during the session
completed	BOOLEAN	Whether the session was completed

titles

Column	Type	Description
title_idPK	INT	Unique title identifier
title_name	VARCHAR(100)	Netflix title name

Your question is Binge-Watching Pattern Mining. Start with the requirements and the two tables in the Question tab.

Run and submit as often as you like. When you're ready, talk me through your approach or go straight to the code.

Problem