The professional speaking world, as structured data

Mezzo is a continuously updated reference dataset of who speaks where, when, and with whom - across the marketing, brand, design, and ideas speaking circuit. Every speaker resolves to a single canonical profile with a verified speaking history, topics, and the events where they appeared. Think IMDb for the speaking world.

What's in it

• Speaker profiles - canonical identity, topics, and full speaking history, cross-source entity-resolved.
• Talks - titles, the event and year given, and video where available.
• Events - conferences and their lineups, by edition and city.
• Shared-stage connections - 8,964 verified pairs of speakers who appeared at the same event edition (the co-appearance graph). This is the count of genuine stage-sharing links, not broad same-event attendance.

Coverage 1984-2026, updated continuously.

How to cite

Access & licensing

The dataset is free to browse and cite with attribution (CC BY-NC 4.0). Individual records live at their canonical pages - start from the speaker directory or the events index.

Commercial use, bulk export, or an API of the full graph? That's a licensing conversation - get in touch.