Event Intro
How EventIntro Works
EventIntro turns a five-question survey into a personalised set of matches for every attendee. You set up a cohort, members complete a survey when they join, and an LLM-driven pipeline produces matches, breakout-group assignments, and conversation starters before every event. Members chat during and after via the platform's built-in messaging.
Step 1: How do you create an EventIntro cohort?
Sign up at the registration link, click "Create New Community" from your dashboard, give it a name and description, optionally upload a banner, and set basic event parameters (date, time, format). The wizard pre-fills sensible defaults for everything except the name and description; the cohort is the persistent home for your members across events.
The cohort is the unit of subscription and the home for member profiles that persist across events. Think of it as a year-round hub — events live inside it, not as standalone things. For a deeper walkthrough, see the Event Host Guide.
Step 2: How does the EventIntro survey work?
Each new member completes a five-question survey when they join the cohort. The questions ask about goals, challenges, offerings, professional interests, and anything else they want to share. The questions are fixed across every cohort — that consistency is what lets the matching pipeline produce reliable seek/offer keywords from every response.
The survey shape — explicitly asking what each person can offer and what they're seeking — is what lets the matching pipeline produce complementary matches rather than just lookalikes. The questions are not host-customisable today: signal consistency drives match quality, and ad-hoc question changes would degrade it. We've written about the design choices in The 15-Minute Event Setup.
Step 3: How does the AI generate matches?
EventIntro processes each member's survey through an LLM that expands the responses into a richer profile and extracts seek/offer keyword pairs. Those keywords are converted into vector embeddings and stored in a pgvector index. When you create an event or breakout, the platform queries the index for each attendee's complementary fits and balances groups for diversity, not just topical similarity.
The pipeline is deliberate about complementary matching: person A's offerings are matched against person B's challenges, and vice versa, instead of pairing people who already share interests. For background on why this works, see The Science of Small Groups.
Step 4: What happens during and after the event?
Each attendee gets a personalised "top picks" page with their best matches and an LLM-generated introduction for each. Breakout groups are formed automatically. After the event, members can continue conversations via the in-platform chat; facilitators get engagement metrics and can run follow-up nudges to keep cohort momentum between events.
Live-event tooling — vibe-based regrouping, mid-event notifications, engagement signals — is documented in The Facilitator's Guide to Reading the Room.
Walk the wizard yourself
The fastest way to evaluate EventIntro is to create a cohort and step through the in-product setup. About fifteen minutes end-to-end.
Get started