Run only as many variants as the hypothesis requires

If it's possible to test a hypothesis with a single variant, don't test additional variants.

Running fewer variants lowers your Minimum Detectable EffectMinimum Detectable Effect
Minimum detectable effect (MDE) is a statistical characteristic of each experiment you design.

This number indicates the percent change that must be caused by one of your variants in order for the...
and simplifies implementation and QA.

This means you'll run more tests, and more of your tests will yield conclusive results. Which is vital, because 80% of your ideas won't work80% of your ideas won't work