RSVP Movies: SQL Strategy Analysis
Turning an IMDB-scale database into a genre and talent investment plan
Problem
RSVP Movies, a fictional production house, needs a data-backed answer to a very real question: what should we make next, and with whom? Rather than greenlighting projects on instinct, this project queries a relational, IMDB-style database — movies, genres, ratings, directors, and cast, across roughly 8,000 films and 25,000+ people — to ground genre choice, casting, and partnership decisions in actual performance data.
Approach
Working across six linked tables (movies, genre, ratings, names, director mapping, role mapping), the analysis moved through four segments of increasing complexity: basic table-shape and null checks, aggregation and filtering to size up genres and years, subqueries to isolate top performers, and multi-table joins to connect a movie's genre and rating back to the specific directors and actors behind it.
Key Findings
4,285
Drama titles — the single largest genre by volume1,078
Drama releases in 2019 aloneTop 3
Drama, Action, and Thriller lead every year sampledGenre volume and rating data were then cross-referenced with director and cast performance to separate three distinct strategies: regionally-proven talent for local hits, globally-recognized studios and stars for international reach, and multilingual-friendly production companies for cross-market appeal — rather than one generic recommendation.
Strategic Output
Genre: lead with Drama, backed by Action and Thriller as a diversification play.
Partnerships: different production partners recommended depending on whether the goal is regional, global, or multilingual reach.
Talent: director and cast shortlist derived from historical ratings and genre fit, not name recognition alone.
Roadmap: immediate, mid-term, and long-term moves, so the recommendation reads as a plan rather than a list of facts.
Why It Matters
The SQL itself was the easy part — the real work was translating query output into a recommendation an actual decision-maker could act on. That's the same muscle I use turning audit and compliance data into recommendations for stakeholders: numbers only matter once someone can decide something because of them.