Traffic allocation
Traffic allocation controls how many visitors take part in an experiment and how they're split across your variations. It works at two levels:
- Experiment traffic — the share of eligible visitors who enter the test at all (for example, 50%). The rest see your site unchanged.
- Variation split — how the entering visitors are divided across the control and each variation (for example, 50% / 50%, or 60% / 40%).
These two multiply. If the experiment traffic is 50% and the split is 60% / 40%, then of all eligible visitors, 30% see variation A and 20% see variation B — the other 50% never enter the experiment.
Example
| Setting | Value | Share of all eligible visitors |
|---|---|---|
| Experiment traffic | 50% | — |
| Variation A (control) | 60% | 30% |
| Variation B | 40% | 20% |
| Not in experiment | — | 50% |
So a 50% experiment with a 60 / 40 split shows variation A to 30% of eligible visitors and variation B to 20%, while the other half see the site unchanged.