Random Group Generator

What this generator does

Divides a list of people into random groups, sized by how many people each group should hold at most; the random team generator takes the number of groups instead. It is the same underlying split as the team generator, framed for classroom and workshop grouping rather than for sport.

How to use this tool

1. Enter participant names, one per line.
2. Choose your target group size.
3. Generate to create random groups.
4. Reshuffle any time for new groups.

Understanding the controls

People per group

From 2 to 50, and the most any group will hold. The page makes as many groups as that needs and deals the shuffled names round them, so group sizes differ by at most one person: 26 people at 6 a group are five groups of 6, 5, 5, 5 and 5, not four sixes and a pair.

Participants

One name per line; blank lines are ignored. The list can be saved to this device from the panel under the box.

Worked examples

19 people in groups of 5

Four groups, of 5, 5, 5 and 4 — as many groups as it takes to keep every one at five or fewer, with the shortfall in one group rather than a leftover on their own.

26 people in groups of 6

Five groups, of 6, 5, 5, 5 and 5. The size you give is the most a group holds: the list is dealt round as many groups as that needs, so the groups come out within one of each other rather than four full ones and a pair left over.

Common use cases

• Workshop or classroom breakout groups
• Group project assignments
• Discussion circles at events

Randomness and fairness

The list is shuffled once with your browser's cryptographic random number generator and then dealt into groups in order, so every person is equally likely to end up in any group and the order you typed the names in makes no difference to the result.

For how randomness is produced across the whole site, see how Generate Random works.

Limitations and good to know

• Group size is the only thing being balanced. Discussion groups usually also care about mixing who has and hasn't spoken yet, or spreading out people who already know each other — none of that is modelled here.
• Running it again for the same room produces an independent new split, with no memory of who was grouped with whom last time.
• The participant list can be saved to this device and restored next time, but the grouping itself is not kept — copy the groups before you leave if you need that exact split again.

Common mistakes

Expecting groups to be mixed by ability or to keep particular people apart

The split is random and nothing else — the tool knows nothing about the people on the list. If certain pairings need to be avoided, the group constraint solver takes must-be-apart and must-be-together rules and says exactly why when they cannot all hold.

Privacy and your data

Names are grouped entirely in your browser, are never uploaded, and never appear in the page address. They are stored only if you press Save in the panel on this page, and then only in this browser, where the same panel deletes them.