How to avoid repeat pairings in workshops
A good rotation is planned as one schedule, not as a stack of independent shuffles. Start with the number of conversations each person can realistically have, then add only the workshop rules that matter enough to reduce that freedom.
Guide visual
Fresh table rotations
Repeated rounds with fewer repeated pairings.
Decide what a successful rotation means
“No repeats” is only one possible goal. Write down the priorities in order before generating anything.
- Choose the number of rounds and the preferred group size from the activity format, not from a desire to force a perfect schedule.
- Decide whether every participant should meet new people, whether roles should be spread, or whether both matter.
- Identify genuine requirements such as fixed facilitators or prohibited pairings. Everything else should remain a preference.
- Plan for attendance changes. A schedule built for 24 people no longer has the same repeat properties when 3 people are absent.
Build the complete rotation together
Generate all workshop rounds in the same scenario so later rounds can be evaluated against the earlier ones.
Clean the participant list
Use one row per person and remove duplicates. Add an attribute only when it changes a real facilitation decision.
Set groups and rounds
Choose people per group when conversation size matters; choose a group count when rooms or table hosts are fixed. Then set the complete number of rounds.
Enable repeat minimization
Keep the rounds in one solve. Generating each round separately discards the encounter history that the objective needs.
Add requirements sparingly
Fix hosts or add together/apart rules only when violating them would make the workshop unusable. A keep-together rule deliberately creates repeated contact.
Generate and inspect
Check capacities and requirements first. Then inspect repeated pairings and whether one participant carries more repeats than the rest.
Revise the model, not just the roster
If the result is poor, remove an unnecessary rule, change group size, or reduce rounds. Manual swaps can repair one pair while creating several others.
Read the repeat pattern, not just the score
- The Pair Meeting Histogram shows the shape of repetition across the whole workshop. “Repeated” counts distinct pairs that shared more than one session, while the bars reveal whether those repeats stop at two meetings or keep accumulating.
- The Pair Meeting Matrix explains the outliers behind that distribution. A selected cell names the pair, lists their shared sessions, and shows pair-local objective cost and together/apart context.
- A high count with required-together context is an intended consequence of the scenario. A high count without that explanation is a useful reason to compare a shorter rotation, another group size, or a version with an unnecessary rule removed.
- Unique Contacts and Repetition Penalty help compare candidate schedules only when roster, groups, rounds, and scoring settings are identical. Across different setups, the totals answer different questions.
- The session rosters under Your groups remain the operational truth: they are what the event producer should inspect and export.
Continue in GroupMixer
Product links open the corresponding section of your current scenario. The Help topic explains how to interpret the resulting evidence.
Worked example: four workshop rounds
The embedded scenario has 24 people in groups of 4 across four rounds. Each person has 12 contact slots, fewer than the 23 other participants, so the basic arithmetic does not force a repeat. The added rules still matter.
Setup facts
- 24 participants, groups of 4, 4 rounds
- roles balanced across tables where possible
- Maya Chen and Owen Brooks required together
- two prohibited pairings
Review the result
- Maya and Owen must repeat each other because the together rule spans the rotation.
- Check whether role balance displaced fresh contacts elsewhere.
- Remove the together rule if it is only a preference, then compare the repeat distribution.
Try this setup in GroupMixer
This tool is preloaded with the example from this guide. You can edit the participants, constraints, sessions, and balance settings before generating groups.