Data Story

The Corporate Family Tree

A paper genealogy of corporate affiliations: branches, staples, and film strips hanging from the family.

filmimdbtmdbcompaniesownershipnetwork
Dataset scope
52503
companies
3260
edges used
24
families
Parent/Sub
filter
Affiliations lack dates. This story avoids causal language and frames the tree as a navigation layer for corporate structure.
Loading corporate family tree
Hypothesis

Larger corporate families produce larger slates; whether that implies homogenization depends on genre diversity within the family.

Question: Do larger corporate “families” produce more diverse slates or more homogenized ones?

Method: Build connected components under Parent/Subsidiary edges; aggregate film outputs; compute genre entropy and budget/revenue medians.

Prediction: Large families touch more films; diversity varies and is not guaranteed to increase with size.

Test: Compare family size vs entropy and financial medians; inspect families as navigation layers rather than causal claims.

Narrative Arc
Act I

Affiliation edges grow into families — legal structure turned into a visual tree.

Act II

Branches unfold like paperwork; staples animate to reveal local structure.

Act III

Film strips hang from the family: diversity becomes a measurable fingerprint (entropy), not a vibe.

Datasets
  • imdb.company_affiliations
  • imdb.companies
  • imdb.film_companies
  • tmdb.movies
  • 18_corporate_family_tree.json
Limitations
  • Affiliation edges are incomplete and may contain errors.
  • No dates: consolidation is inferred, not observed over time.
  • One weak edge can merge families incorrectly.
Next

Want another story? Head back to the film data stories index.

Back to indexarrow_forward