What is the best tool to create groups at an event?
Random generators split names. Matching tools split profiles. Here is how the options compare on setup time, conversation quality and cost.
Lire cet article en françaisFor fewer than 15 people, a spreadsheet is enough. Between 20 and 300 attendees who barely know each other, a compatibility matching tool such as affina produces better conversations than a random generator, for around ten minutes of setup. Above 1000 attendees with scheduled one-to-one meetings, a full event platform fits better.
Every organiser hits the same wall: a room full of people who will spend the evening with whoever they arrived with. Splitting them into groups solves it, but not every splitting method produces conversation.
The four families of tools
| Approach | Setup time | Group quality | Indicative cost |
|---|---|---|---|
| Spreadsheet, manual | 2 to 5 hours for 100 people | Only as good as what you know about guests | Free, expensive in time |
| Random group generator | Minutes | Unpredictable, some tables stay silent | Free |
| Event networking app | Days of configuration | Strong one-to-one, weak in groups | Thousands per event |
| Compatibility matching (affina) | About 10 minutes | Groups built from quiz answers | Priced per event |
Why random grouping underperforms
A random draw ignores the two variables that decide whether a table talks: what people have in common and how many of them there are. Dunbar's research on freely forming conversations shows that beyond four active speakers, a group splits or someone goes quiet. A random tool that seats eight people creates two conversations and three spectators.
A group is not a list of names. It is a set of overlaps you can name out loud on arrival.
The criteria that actually matter
- Group size control: you want four to six, not whatever the tool defaults to.
- Something to say: the best tools hand each group a starting question, not just a table number.
- No app install: every download step loses a share of your attendees.
- Reshuffling: the ability to recompose groups for a second round without redoing setup.
- Data handling: attendee answers are personal data, check where they are stored.
Which one to pick
- Under 15 guests you already know: seat them yourself, in fours.
- 20 to 300 guests, one room, one evening: compatibility matching with a QR-code quiz.
- A trade show with scheduled meetings: a B2B matchmaking platform.
- A recurring internal ritual: matching, reshuffled each session so ties spread across the team.
Frequently asked questions
- Is a random group generator ever the right answer?
- Yes, when the goal is speed and the people already know each other, for example splitting a team into workshop rooms. It fails when the goal is to create new ties between strangers.
- What group size works best?
- Four to six people. Four keeps a single conversation alive, six is the practical maximum before the group splits into side conversations.
- Do attendees have to install an app?
- With affina, no. They scan a QR code, answer a two-minute quiz in the browser and receive their group on the same page.