I built a single power scale for people, companies and institutions (what worked, what broke)

A solo builder from India has created ApexHierarchy, a platform that ranks individuals, companies, institutions and countries on a single, unified scale. The site categorizes every profile into one of eight tiers, ranging from "Town" to "Global", with each tier containing Low, Mid, High and Top positions. The project is still in its early stages, but the creator invites honest feedback.

The platform allows users to search for any person, company, institution or country, and it automatically researches public information about the subject. Six key dimensions are used to rank each profile: Net Worth, Reach, Power, Influence, Network and Legacy. Additionally, a rising or falling trajectory is provided for each of these dimensions. A connections map feature allows users to explore the relationships between different profiles, and a network graph displays each profile's connections.

The site also includes several unique features, such as the Apocalypse Index, which ranks entities with the power to reshape the world order, and global rankings for net worth and institutions. The platform covers 40+ cities, including Mumbai, Chicago, Paris, and Tokyo, and provides net worth rankings for India, the US, and China, as well as the richest people in each of these countries. There is also an A-Z directory containing 777 profiles.

Every profile on the site lists its sources, marks which are corroborated, and shows a data-confidence score. Anyone can report a mistake, claim their own profile, or submit a missing one. The platform is free to browse, with paid plans starting at ₹499/month or $12/month outside India. These paid plans offer more profile views, PDF export, a watchlist, history, and a recent-developments feed.

The process of creating each profile involves gathering information from public sources, using an LLM to write the text and pull figures, and then assigning the tier using a formula rather than the LLM's judgment. This ensures that the tier assignment remains objective and based on the data rather than the model's output. A hard reach-based cap prevents a profile from being ranked above its actual public reach, regardless of any claims it may make about itself.