EventIntro FAQ

EventIntro is a small-event networking platform for cohorts of 10–100 people. It runs an AI-driven matchmaking pipeline on top of a five-question survey and produces personalised introductions, breakout-group assignments, and follow-up chat.

EventIntro is built for facilitators of small, high-context groups — cohort leaders running ongoing communities, alumni-association event organisers, mastermind hosts, and corporate learning teams running cross-functional offsites.

EventIntro is not for large public conferences (500+ attendees), ticket-driven mass events, or one-off trade shows. We don't sell tickets, we don't host streaming sessions, and we don't compete on event-management feature breadth with Eventbrite, Hopin, or Bizzabo.

EventIntro takes the matchmaking work seriously. Most networking tools either skip matching (relying on chance) or do shallow keyword-tag matching. EventIntro asks each attendee five short questions, uses an LLM to expand the answers into a richer profile, and pairs people whose offers match each other's needs.

EventIntro is a new platform, in active build. The matching pipeline runs end-to-end; we don't yet have aggregate customer-outcome data to publish. As we have data we can share without breaching attendee privacy, we will.

The in-product setup wizard takes about fifteen minutes for a default configuration: roughly three minutes to create the community, two for the survey, five for invitations, three for AI options, and two for preview-and-launch.

Not today. The five questions are fixed across every cohort because the matching pipeline (LLM enrichment, seek/offer keyword extraction, embedding-space search) is tuned for the specific shape of those questions. Custom questions would introduce signal drift and degrade match quality, especially across cohorts where members carry their profile between groups. You can customise the cohort name, banner, invitation email, breakout-group size, and matching priority — just not the survey itself.

Yes. EventIntro supports in-person, virtual, and hybrid events. It works alongside Zoom, Microsoft Teams, and Google Meet. The platform doesn't host video itself — it sits above whatever video tool you use and handles the matching and follow-up layers.

EventIntro sends automated reminders and shows profile completion as a percentage on the host dashboard. You can also require complete profiles for matches to surface, which gently pushes laggards to finish before they get any matches at all.

EventIntro is built for 10–100 attendees. Strategic discussion among peers fits 8–15; mixed networking and short-form learning fits 20–40; community-building runs well at 50–80. Past 100, attendees default to existing connections without strong structural support.

Create the event at least one week in advance to give members time to join and complete their profiles. Two weeks is better — pre-event engagement (members reading their match list, exchanging messages) is a leading indicator of conversation quality on the day.

EventIntro processes each attendee's survey through an LLM that expands the answers and extracts seek/offer keyword pairs. Those pairs are converted to vector embeddings; matching queries the index for each attendee's complementary fits — people whose offerings answer their challenges, and vice versa.

EventIntro supports multiple LLM providers — Google Gemini, OpenAI GPT, Anthropic Claude, and Groq — with automatic fallback between them on failure. The match quality depends more on the survey design and the embedding pipeline than on which underlying LLM is selected.

Yes. The host dashboard shows a preview of all matches before the event. You can override pairings or breakout assignments before the event goes live. The AI is a productive default, not a replacement for the facilitator's judgement.

By default, complementary — EventIntro pairs people whose offerings answer each other's challenges. You can switch to similarity-based matching if your event is built around shared interest (a study group, say) rather than mutual help. Most events benefit from complementary matching.

Update your profile with more specific information about your goals and offerings. The matching pipeline reflects what you write; if your profile is generic, your matches will be generic too. Facilitator feedback also flows back into how matches are scored.

EventIntro uses AES-256 encryption for data at rest and TLS for data in transit. Communities and events are private by default — only members of a specific community can see each other's profiles. Profile data is owned by the attendee, not the host.

Yes. Hosts can export attendee data and connection reports from the dashboard. Attendees can edit or remove their profile information at any time. Data retention is host-controlled.

No. EventIntro communities are invitation-only. Profile pages are not publicly indexed. The robots.txt blocks every authenticated route, including profile and conversation paths.

EventIntro uses third-party services for delivery (email, hosting, LLM providers); attendee data flows through those services as needed for the product to work. We don't sell data, and we don't use attendee profiles to train any third-party model.

EventIntro is a single annual fee per community ($500/year), covering unlimited events. There are no per-attendee or per-event charges. See the pricing page for the full set of features included.

You can join the EventIntro user community to experience the matchmaking yourself, and request a 1:1 demo at any time. The annual subscription unlocks the host-side features.

Each cohort has its own annual subscription. You can manage multiple cohorts from one host account, but the pricing is per-cohort to keep the model predictable.

Every account gets 1:1 onboarding to walk through community and event setup. Ongoing support is via email at hello@eventintro.com plus video tutorials. We don't have a 24/7 ticket system — events are scheduled, so support is concierge.

Zoom (breakout-room assignment), Microsoft Teams, Google Meet, Slack (notifications), and CRM-style data exports. The platform isn't trying to be the centre of your tooling stack — it's the matchmaking layer above whatever you already use.