Share a list of tasks out among a list of people at random, either dealt evenly so workloads differ by at most one, or drawn independently. Good for chores, rotas and team duties where nobody should be choosing.
Settings
Recent results
Generated locally in your browser β your settings and results never leave this page.
Random and fair are not the same word
Ask for tasks to be handed out at random and most people picture an even split. Genuine independent assignment does not produce one. Give nine chores to three people by drawing a name for each, and the most likely single outcome is a 4-3-2 split; 6-2-1 is entirely ordinary; 9-0-0 is rare but perfectly possible.
Everyone had the same chance at every step. Nobody was disadvantaged. And the result still looks unfair to whoever ended up with six, because what people want from “assign these randomly” is usually an even distribution arrived at impartially β not statistical independence.
Even sharing gives them that. Shuffle the people, deal the tasks round the circle, and the counts can differ by at most one. Which tasks land where is still entirely down to the draw; only the workload is constrained.
Where each mode belongs
Even sharing is the right default for anything real: household chores, team duties, classroom jobs, support rotas. The point of drawing lots for these is to remove favouritism from the decision, not to produce a lopsided outcome that then has to be argued about.
Independent assignment is the honest choice when you are modelling arrival rather than organising work β simulating how tickets land on a team, testing whether a queue copes when the distribution is uneven, or demonstrating what independence actually looks like. The tool warns when independent mode produces a noticeably lopsided result, because that is the case most likely to be unintended.
Uneven numbers
Tasks and people rarely divide neatly. With eleven tasks and four people, three get three and one gets two β that is as even as it can be. The tool reports the spread so you can see at a glance whether the split is acceptable.
More people than tasks is the case worth watching, because somebody genuinely receives nothing. That is fine when it is intentional and awkward when it is not, so it is stated rather than left to be noticed.
Randomness settles arguments it cannot settle twice
The real value of drawing lots for chores is not statistical. It is that nobody chose. A rota produced by a draw removes the suspicion that the person organising it gave themselves the easy jobs, and it removes the negotiation that otherwise fills the time.
What it does not do is stay fair across repetitions. Each run knows nothing about the last, so the same person can draw the worst job several weeks running β and will, occasionally, exactly as often as probability predicts. Over a term or a quarter that stops feeling like chance.
If long-run fairness is what you actually need, rotation beats randomness. Draw the order once, then rotate it each period. Everyone gets every job in turn, the process is still impartial, and nobody can end up with the same task three times by accident.
Task order is not priority
Tasks are shuffled before dealing, so the order they appear under a person’s name is arbitrary. Reading it as a sequence to work through would be a mistake. If some jobs must happen before others, that is a dependency, and dependencies should be decided deliberately rather than inherited from a shuffle.
Privacy
Names and tasks stay in your browser. Nothing is uploaded, stored or included in the page link.
Tasks that are not equal
Even sharing balances the number of tasks, not their difficulty. Four jobs each is only fair if the four jobs are comparable, and in most real rotas they are not β one chore takes five minutes and another takes an hour.
The practical fix is to split the heavy tasks into several lines so they carry proportionate weight in the deal, or to run separate draws for heavy and light jobs so each is shared evenly on its own terms.
What does not work is assuming the count alone represents fairness. A rota where everyone has three tasks and one person has all three difficult ones will be argued about, and correctly so.
How to use the Random Task Assigner
- List the people, one per line.
- List the tasks, one per line.
- Leave even sharing on unless you want each task assigned independently.
- Assign, then copy the result or the version grouped by person.
Frequently asked questions
What does even sharing do?
It deals the tasks round-robin over a shuffled list of people, so the number of tasks each person receives differs by at most one. Turning it off assigns every task independently, which is more genuinely random and frequently very unequal.
Why would I ever turn even sharing off?
When you are modelling something rather than organising it β simulating how work arrives at a team, for example. For an actual rota, even sharing is almost always what people mean by "assign these fairly".
What happens if there are more tasks than people?
People receive several tasks each, spread as evenly as the numbers allow. With seven tasks and three people, someone gets three and the others get two.
What if there are more people than tasks?
Some people get nothing, and the tool says how many. That is a legitimate outcome β it just should not be a surprise.
Is the order of tasks within a person meaningful?
No. Tasks are shuffled before being dealt, so the order shown is the draw order rather than a priority. If the sequence matters, decide it separately.
Can I run the same rota every week without repeating assignments?
Not automatically β each run is independent and has no memory of previous ones, so the same person can get the same chore twice running. For rotation rather than random assignment, put people in a fixed order once and rotate it each week.
Do the names and tasks leave my browser?
No. Both lists are processed entirely in the page. Nothing is uploaded to a server, written to storage, recorded in analytics or placed in the page link, and closing the tab discards everything. That matters most when the list is a team rota or a household, where the names identify real people.
