Carta Ownership Map

TL;DR

Led design for Carta's first multi-entity ownership map, giving PE-backed group admins a real-time visual of their full ownership structure and support to drill down into reporting and information that matters. This feature so far is used across 10K+ multi-entity companies on the platform.

Overview

Carta is the leading equity management platform — 50,000+ companies, $4.5T+ in assets, 9,000+ funds across 160+ countries. It helps private companies manage cap tables, issue equity, run valuations, and administer funds.

Problem

Carta was built around single entities. For PE-backed portfolio companies managing 6–20+ entities across multiple jurisdictions, there was no view above the individual cap table. The core problems:

• No structural view. Other than a flat account switcher, there was no way to see how entities related to each other. Admins pieced together their ownership structure manually using spreadsheets alongside Carta.
• No way to verify data accuracy. Carta held the data but there was no mechanism to confirm it matched the legal structure on paper. The only path was drilling into each entity's cap table one at a time.
• No visibility into cascading effects. A termination, a new grant, a share class change meant that admins had no way to see how an action in one entity rippled across the group.

Approach

Three principles guided the design:

• Communicate relationships. Show the structure of who owns what, how they connect.
• Progressive disclosure. Primarily focus on the global view on the canvas with the ability to drill down into more in-depth analysis on entities and ownership data.
• Keep admins in context. Allow users to isolate a specific ownership chain without losing their sense of the full structure around it.

Solution

An interactive ownership map that renders the full multi-entity structure as a live canvas, showing funds at the top, flowing through holding companies, operating entities, management vehicles, and down to stakeholder pools. Five node types cover the full range of actors in a PE structure: Fund, Entity, Investor, Stakeholder, and Trust.

Three interaction states keep the experience focused. Hover highlights a node's direct connections and dims everything else. Select centers the node on the canvas with a smooth pan-and-zoom animation and opens a detail drawer.

The AI agent handles what the canvas can't. Admins query scenarios directly: "What happens to ownership percentages if this employee becomes fully vested?" or "Which entities are affected if we terminate the Senior Leadership Pool today?" The agent traverses the ownership graph, models the outcome, and highlights the affected nodes on the canvas.

Key learnings

• The map is only as useful as the depth it enables. Surfacing structure is the starting point. The bigger opportunity is helping admins navigate the ownership chains that matter to their specific role, not just rendering all of them equally.
• The AI agent supports interpretation, not the other way around. The agent is only as useful as the map beneath it. Admins need to read the structure independently first and the agent extends that by modeling scenarios and surfacing deeper detail. The map must communicate clearly on its own.