A random seating chart that actually mixes people up

Leave seating to chance and it isn't really chance at all. People sit with the friends they walked in with, the loud table fills first, and the two people who'd actually have hit it off end up at opposite ends of the room. Whether it's a classroom you want to stop forming permanent cliques, a wedding where the families need gently separating, or a team lunch meant to break the usual silos, the point of a seating chart is to override the default. A random one does exactly that — and does it in a way nobody can accuse you of engineering.

Paste your guests, set a table size, done

The random seating chart works from the simplest possible input: your guests or students, one name per line. Set how many seats a table holds in the Options panel, click Assign tables, and you get a clean grid — Table 1, Table 2, and so on — each filled with a random handful of names. There's no drag, no spreadsheet, no fiddling with a floor plan. It's the thirty-second version of a job that usually eats an evening, and because the assignment is genuinely random, the result carries a fairness a hand-drawn chart never can: you didn't put anyone anywhere on purpose.

"Seats per table" is a maximum, not a promise of even tables

This is the one detail worth slowing down for, because it's where charts go wrong. The seats number is the most people any table will hold, not a guarantee that every table is equal. Under the hood the tool shuffles your whole list fairly, then deals it out in chunks of that size. If the number of guests divides evenly, you get tidy, equal tables. If it doesn't, the leftover people form one final, smaller table — and in the worst case that's a single guest sitting alone.

A quick example: 18 students with a table size of 6 gives you three neat tables of six. But 20 students at the same size of 6 gives you 6, 6, 6, and then a fourth table of just 2. Twenty-five guests at a size of 6 leaves one person at a table by themselves. The fix is to pick a size that divides your headcount — for 20 people, a size of 5 gives four even tables of five — or to glance at the last table and merge a straggler into another. Knowing the remainder lands on the final table is the difference between a chart that looks intentional and one that accidentally exiles someone.

A fair shuffle, so no seat is "the bad seat"

The mixing itself uses a proper Fisher-Yates shuffle on your full guest list before the tables are carved out, which means every person is equally likely to land anywhere. There's no quiet bias toward whoever you typed first, no clustering of names that happened to be adjacent in your list, and no front-of-room versus back-of-room favouritism. When a student grumbles that they always get stuck at the worst table, you can say — truthfully — that the assignment didn't know or care which table that was. That even-handedness is most of the reason to randomize in the first place.

Reshuffle freely, or lock it in with a seed

Most of the time you'll just want a fresh arrangement, and clicking again gives you one — handy for rotating a classroom weekly so the same neighbours don't calcify into a clique. But there's also an optional Seed field, and it's quietly powerful. Leave it blank and every draw is unpredictably random. Type any word or number into it and the chart becomes reproducible: the same guest list with the same seed always produces the same tables. That's how you publish a wedding plan in advance and regenerate the exact same one on the day, or prove to a parent that this week's seating wasn't hand-picked — you can replay it from the seed in front of them.

Private, offline, and easy to hand out

The whole thing runs in your browser. Your guest list never leaves your device and isn't uploaded anywhere, which matters when the names belong to children or to a private event. It also keeps working with no connection at all — useful when you're setting up in a venue hall with no wifi. When the chart looks right, the result comes back as a plain list you can copy, and you can export it as a CSV with a row per guest and their table number — exactly the format to drop into place cards, a check-in sheet, or a printed plan for the door.

When a different tool fits better

Seating by table size is the right tool when you care about where people sit. If you only need to split a roster into a set number of even groups — five project teams rather than tables of a fixed size — a team / group divider is the cleaner fit. If you're running sessions over time and want to actively avoid pairing the same people twice, reach for the no-repeat group mixer instead. And for splitting a video call into rooms, the breakout room assigner speaks that language directly.

Seat the room, then forget about it

A good seating chart is invisible — people just find their table and the mixing happens on its own. Paste your names, choose a table size that divides cleanly, glance at the last table to make sure nobody's stranded, and let the seating randomizer do the awkward social engineering for you. Add a seed if you want to lock the plan in, and you've turned a fraught little task into a single click.