The State of Event Attendee Intent

The finding that changed how we match people

At one regional business forum, 60% of profiled attendees said they came to sell: to find clients, generate leads, win contracts. Not to learn, not to buy. That fact appeared nowhere in the registration data, and the event's networking programme had been designed on the assumption that attendees came to meet each other as peers.

It matters more than anything else we have found across seven deployments, because almost every networking format in the industry assumes a room of buyers and learners. When the room is majority sellers, matching people by similarity does not produce good meetings. It produces competitors at the same table. That is what happened, and an attendee told us so in plain language after he got his schedule.

The proportion is also not a constant. At a developer-focused tech conference the same measurement returned about 15%. An organizer cannot borrow either number. They have to measure their own, and registration will not tell them.

Buyers, sellers and learners, by event type

The mix moves with the event's format and audience, and the two ends of our sample are far apart.

Four times the concentration of sellers, in the same product, in the same year. This is the number an organizer cannot borrow from anyone else.

Regional SME business forum: 60% client acquisition. The audience was largely small business owners and independent consultants (HR advisors, insurance brokers, IT providers, coaches, lawyers), most of whom sell professional services to the same buyer profile. A speed-networking format in that room seats sellers with sellers unless something actively prevents it.

Developer-focused tech conference: 9 of 58 identified attendees, about 15%, stated sell-side intent outright, including representatives of three sponsors. One attendee said plainly that he did not want to be sold to. Both groups were legitimate attendees, and both were meeting in the same undifferentiated directory.

Multi-stage tech festival: a different pattern again. 78% of conversations were navigation rather than networking: which stage, which room, what is on next. With five concurrent stages, the dominant attendee need was orientation, and networking intent surfaced only after that was satisfied.

The practical consequence is that "sponsors want to meet attendees" is not a plan. At the first event, a sponsor selling to SMEs was surrounded by qualified prospects. At the second, sponsors were competing for a minority of the room while a majority of self-identified sellers looked for the same buyers the sponsors wanted. Those two events need different sponsorship products, and neither organizer could have known which they had from their registration export.

How many attendees will hold a real conversation

Of attendees who engaged at all, 61 to 69% completed full discovery, meaning they described their work, their goals and their interests, not just asked a question. The written-survey baseline organizers compare against is 10 to 20%.

At the regional forum, 69% of attendees completed full discovery across 151 conversations analysed; median conversation length was 5 messages and 55% ran to 5 or more. At the multi-stage festival, 61% completed discovery across 126 conversations, notable because most of those conversations started as navigation questions. Intent surfaced as a by-product of someone asking which stage to go to.

Nobody sat down to fill in a profile. The discovery completed inside conversations people entered for other reasons, which is why the completion rate is several times the survey baseline. A form asks for effort in exchange for nothing; a conversation answers a question first.

One boundary on this number: it is a rate over attendees who engaged, not over attendees who registered. The next section is about that gap, and it is the larger number.

What drives engagement, including where it failed

One variable dominates: whether the organizer put the opt-in on the registration form and then promoted it. Not the channel, not the copy, not the technology.

We publish the 19% and the 6% because a report in which everything worked is marketing. The read-versus-reply split in the middle case is the most useful detail: two-thirds of attendees opened the message and only a third answered it. Attention was not the constraint. The message had not made its value obvious enough to earn a reply, and at a content-led event it was competing with the programme itself.

For organizers evaluating any attendee-engagement product, this is the number to interrogate. Ask for engagement as a count over registered attendees at a named event. Rates over "active users" are not comparable to anything.

Method, and what this cannot tell you

Intent was captured in opt-in conversations on WhatsApp and web chat, before and during each event. Transcripts were analysed per event and aggregated into segments; delivery, read and reply counts come from platform logs. Every figure on this page is a count over a stated base from a single named deployment. Nothing is averaged across events.

Limitations, stated plainly

• This is a convenience sample, not a representative one. These are seven events that chose to deploy a conversational agent. They skew B2B, skew toward organizers already worried about networking outcomes, and are mostly European. Do not read these percentages as industry norms.
• Intent segments come from what attendees chose to say. Someone who did not mention wanting to sell is recorded as not having said it, which is not the same as it being untrue. If anything, the 60% figure understates commercial intent, because stating it is socially costly.
• Segments overlap and were assigned by analysis, not self-selection. Attendees did not tick a box marked "client acquisition". Their stated goals were read and grouped, which involves judgement.
• The discovery-completion rates are over engaged attendees, not registered ones. Stated again here because it is the easiest number on this page to quote out of context.
• Sample sizes for the intent breakdowns are small. 128 attendees at one event and 58 at another. They are large enough to show that the two events differ substantially and too small to support a precise figure for either.
• We are not a neutral party. We sell a product that captures this data, and this research makes the case that capturing it is worthwhile. Every number is sourced to a deployment report and stated with its denominator specifically so the reasoning can be checked rather than trusted.
• One deployment publishes no engagement figure at all. The large trade show in the sample ran with a delegation scope and no organizer promotion, and we do not have a defensible engagement denominator for it. It contributes content-scale figures only.

No attendee names, quotes attributable to an individual, employer names or contact details appear in this report. Findings are aggregated at event level and published with client consent.

What an organizer can do with this

Four things, none of which require buying anything from us.

• Add one question to your registration form. "Are you here mainly to learn, to hire, to buy, or to sell?" One field, asked honestly, tells you which room you have. It is the cheapest version of this entire report.
• Stop designing networking for a room of peers. If a meaningful share of your audience came to sell, undifferentiated matching will seat them with each other. Either segment the format or accept the complaints.
• Put any attendee-channel opt-in on the registration form. The gap between 76% and 6% in our data is mostly this decision.
• Demand denominators from every vendor, us included. Ask for engagement as a count over registered attendees at a named event.

Citation and reuse are welcome under CC BY 4.0. If you want a figure clarified, or think we have read our own data wrong, tell us. We would rather correct this page than defend it.

Keep reading

• WhatsApp for event attendees: what works and what breaks: the operational detail behind how this data was collected.
• Prove sponsor ROI beyond badge scans: what this intent data is worth to a sponsor.
• Best event networking apps, ranked by outcomes: how to grade any platform against these questions.