Free tools. No account. Nothing stored.

Random Study Group Generator

Here is the situation. It is 9 p.m. the night before a case study is due, you have thirty-two students on the roster, and you need six tables. Counting names by hand is where you make the mistake — you end up with one group of seven and one group of two, and then somebody has to sit alone, and they are going to remember that.

That is all this tool does. You paste the list, say how many groups you need, press one button, and everyone is placed. No account, no spreadsheet, no names leaving your device.

I built it after watching the same argument in every tutorial I ran. Someone is always left out, and it is never the same person twice, which is a nice way of saying everyone gets cheated eventually. Once the split is automated, the awkward part of the conversation goes away too, because there is nothing to argue about. The computer picked it.

Paste the class list, say how many groups you need, and press one button. Groups come out filled evenly, and an optional check tries to keep people from landing in the same group two weeks running.

Names stay on this device. The list and your last round are kept in your own browser storage so a refresh does not wipe them, and nothing is sent anywhere.

What the group maker actually does

It shuffles the names, then deals them out one at a time in a round robin. Round robin is the important bit. If you deal the whole first group, then the second, you get lumpy sizes. If you deal one name to each group in turn, a group of thirty-two into six tables gives you six tables of six, and into four tables gives you four groups of eight. When the numbers do not divide evenly, the sizes differ by at most one. That is the difference between a fair split and an accident.

Four options you can turn on:

  • Fill them evenly — the round robin above. This is the default because uneven groups are the thing students actually complain about.
  • Pick a group at random for each student — every name goes into a group chosen at random. Sometimes a group ends up empty. I left it in because for icebreakers people genuinely do not mind an empty group, and because a lopsided random split is a good demonstration of why the even option is better.
  • Keep these students together — type two or three names in and they are locked into the first group before anyone else is placed. Use it for a pair that works well, or for a support student who should not be the odd one out.
  • Try to avoid the pairings from my last round — remembers the last split on your own device and tries to break up the pairs that were together last time.

About the avoid-repeat option, honestly

This one is a best effort, not a guarantee. The tool builds the groups, counts how many pairs from the last round ended up together again, then tries swapping people between groups to reduce that number. It makes up to six passes and stops when it cannot improve.

Sometimes it gets to zero. Sometimes it cannot, and it tells you how many repeats are left rather than pretending. That happens when there is no arrangement left that separates everyone — with 14 students in 3 groups, each person has to sit with 4 or 5 others, so some pairing from a previous round is going to survive. If you need a real no-repeat guarantee, use a proper round-robin scheduling tool with full history. This one remembers exactly one previous round.

It also only remembers one round. Generate again without ticking the box and the pairs you are comparing against get overwritten with the newest split, not stacked on top of the old ones.

Worked example: 30 students into 5 groups

Paste 30 names, set the number of groups to 5, leave the mode on Fill them evenly. You get 5 groups of 6. Every name used once, no duplicates, no group left short.

Now try 30 into 4. Round robin gives group sizes of 8, 8, 7, 7. That is the best you can do and nobody should be short-handed by more than one person.

Then try 7 into 3 with random mode. You will usually end up with something like 3, 2, 2 — and sometimes 4, 3, 0. Same thirty seconds of work, a completely different feel in the room. That is the point of showing students both.

Tips from running these sessions

Use the real roster, not a memory of it. Students join and leave. The person you remember from week one is not always the person on the sheet in week six, and a spelling variant creates a duplicate who appears in two groups at once. The tool ignores exact repeats of the same name, so a doubled entry does not cost anyone a seat twice — but a typo still becomes two people.

Put one name per line. Commas and semicolons work too, so a list pasted out of a spreadsheet is fine.

Name the groups after something that exists. Tables, rooms, or the four corners of a problem — a case with a strength, a weakness, a risk, and a question. My default label is Group 1, Group 2, which tells you nothing when the sheet is printed and left on a desk.

Rotate rather than re-randomise. If you need real variety over a term, generate once and then shift every person one group along the next time. That guarantees nobody sits with the same people twice without any memory of the previous round.

Do not use this to pick the group that presents. That is a different decision, and it is not random. I have watched a room work out in about ten seconds who the tool is not going to pick, which is fine for a study group and a problem for anything that matters.

Where the names go

Nowhere. The tool runs entirely in your browser, the same as the calculators on this site. There is no account and no server. Two things get stored in your own browser storage on your own device: your class list, so a refresh does not lose it, and the pairs from your last split, so the avoid-repeat option has something to compare against. Nothing is transmitted. Clearing your browser storage for this site removes both, and the Reset button clears the list without touching anything else.

Student names are still names. If you are entering a list you do not have permission to keep, the honest advice is to clear it when you are done, and the Reset button is right there.

Frequently Asked Questions

How do I split a class into groups fairly?

Shuffle, then deal one name to each group in turn until the names run out. That is the round robin method, and it is why the group sizes end up within one of each other. Splitting the list into equal blocks first is the usual mistake, because a remainder of 2 students across 5 groups turns into one group with two extra people in it.

How many groups should I make?

For discussion, three to five is usually the sweet spot. Below three people nobody has to defend a view, and above six the quiet ones stop contributing. For problem sets, aim for two to four people per group so there is a job for each. Then check the size against what you are doing rather than against what divides evenly.

Does it stop me getting the same group twice?

It tries, and it tells you the number when it cannot. Tick the avoid-repeat box and the tool compares your new split against the last one it made on this device, then swaps people between groups to break up old pairs, for up to six passes. When a no-repeat split is impossible with your roster and group count, it reports the leftovers rather than claiming a clean result.

Can I put specific people in the same group?

Yes. Type their names into the keep-together box, separated by commas, and they are placed in the first group before anyone else. Those groups come out larger than the rest, because the rest of the names are still dealt evenly across all of them. It is a useful feature for a pair that works or a student who should not be left as the odd one out, and it does quietly unbalance that group, so use it for two or three names rather than half the class.

Does the tool store student names anywhere?

Not anywhere you cannot see. The list and your last pairing are kept in your own browser storage on your own device, so they survive a refresh and let the avoid-repeat option work. Nothing is sent to a server, there is no account, and no analytics see the names. Reset clears the list from the page, and clearing site data in your browser removes what is stored.

Other tools

Press the button, read the table, and stop spending your evening on arithmetic. That is the whole job.

Related GPA tools