Advanced workflow guide

Build a multi-stage random wheel workflow without mixing the rules

Several wheels can represent a sequence—person, task, time slot, prize—or independent stations. Naming the stages and their dependencies keeps a convenient control panel from turning into an ambiguous chain of random decisions.

For
Workshop producers, teachers, event operators, and facilitators with multi-step draws
Reading time
10 minutes
Reviewed

Step 1

Map each stage and decide whether outcomes depend on one another

Write one line for every stage: its eligible inputs, output, acceptance rule, and what consumes that output. Selecting a presenter and then selecting that person's topic is dependent; spinning room themes and snack choices may be independent. A dependent workflow often requires changing the next stage's eligible list, while independent wheels can remain fixed.

Do not assume that spinning several wheels simultaneously creates a valid combination. If certain people cannot take certain shifts, prizes require different eligibility, or a selected category changes the available prompts, those constraints must be modeled and checked between stages.

Check before continuing

  • Every stage has a distinct question and input list.
  • Dependencies and prohibited combinations are written down.
  • The operator knows when a result becomes accepted.
  • A failed later stage has a defined recovery path.

Step 2

Give each wheel a role, owner, and visible label

Use the available workspace slots—up to four wheels—for stages that the audience can distinguish at a glance. Names such as Stage 1 are weaker than Presenter, Topic, Time Slot, and Bonus because a host can easily spin the wrong active wheel under pressure. Keep unrelated events in a separate workspace or backup rather than overloading one control surface.

Place a short rule in the wheel description or run sheet: whether the stage repeats, removes accepted results, uses weights, or resets for each participant. The active wheel name and entry count should be part of the operator's spoken confirmation before a live spin.

Step 3

Configure mode and weights separately for every stage

A participant wheel may use elimination to produce a no-repeat order, while a topic wheel may stay in normal mode because topics can recur. A prize wheel might use weights based on documented inventory. Review each setting independently; copying one wheel's mode to every stage is convenient but can alter the intended process.

When weights are used, explain what one weight unit means and verify positive numeric values before the session. Configure theme, result sound, mystery labels, and countdown separately as presentation cues for each stage; they never change that stage's population or probability rule. A no-replacement draw changes the future population after each accepted result, so record whether an undo restores the result and when a stage is considered final.

Step 4

Choose active-wheel or spin-all behavior from the dependency map

Spin only the active wheel when the next eligible list depends on the current result, when the audience needs to confirm each step, or when stages have different timing. A spin-all action is appropriate only for genuinely independent wheels whose resulting combination is always valid.

If spin-all is used, define the reading order and acceptance point. Do all outcomes become final together, or can an individual stage fail validation? Avoid selectively rerunning just one unpopular component unless the predeclared rules allow that partial redraw.

Run it in this order

  1. Confirm the named wheels and their entry counts.
  2. Choose sequential or simultaneous operation from the dependency map.
  3. Announce the stage, spin scope, and acceptance rule.
  4. Validate and record results before advancing or resetting.

Step 5

Track history, scores, and transitions without confusing them

History answers what was selected and in what order; a scoreboard counts occurrences; the active list shows who or what remains eligible. These views serve different purposes. In elimination mode, verify the remaining active list after accepting a result. In scoreboard mode, remember that counts describe outcomes and do not force future selections to balance.

For a dependent sequence, copy only the necessary accepted output into the next stage or run sheet. Do not treat an unconfirmed animation result as an input. If an operator mistake is undone, record the correction before continuing so the visible state and event notes remain aligned.

Step 6

Back up the complete workspace and rehearse handoff

Export a workspace JSON backup before a high-effort setup or live event and test its import in a separate browser profile. Each local wheel can hold 1,000 entries, while share links remain limited to 200 entries per wheel and intentionally omit history, scores, and local images. Use the validated full JSON workflow for a large setup and confirm its final expected row after import. Treat both formats as potentially sensitive when they contain participant labels or event details.

Give the backup and run sheet to the designated alternate operator through an approved channel, then rehearse opening the correct wheel and stage. After the event, retain only what the operational or recordkeeping policy requires and clear local data from shared production machines.

Use the right surface

Tools mentioned in this guide

Common questions

Questions about multi-stage wheel workflow

How many wheels can I keep in one workspace?

The advanced workspace supports up to four named wheels. Use them for related stages the operator can distinguish clearly, and separate unrelated events into their own backups or workspaces.

When should I use spin all?

Only when the wheels are independent and every possible combination is valid. Use sequential active-wheel spins when later eligibility depends on an earlier result or each stage needs confirmation.

Can different wheels use different modes and weights?

Yes. Review normal, elimination, scoreboard, and weight settings per wheel because each stage can represent a different selection rule.

Does a share link preserve the whole multi-wheel session?

No. Sharing is intentionally bounded and does not include history or scores. Use a protected workspace JSON backup when complete recoverable state is required.

What should happen if a later stage produces an invalid combination?

Apply the recovery policy written before the session. Record the failed combination, correct the constraint or use the approved redraw scope, and do not selectively rerun outcomes just because they are inconvenient.