Why Change Two Factors Together?
Discover why traditional one-factor-at-a-time experimentation fails to detect variable synergy and how simultaneous adjustments reveal true cross-factor dependencies.

Contents Overview
The Limits of One-Factor-at-a-Time Testing
A common way to study a process is to hold everything constant, change one input, record the result, and then move on to the next input. This one-factor-at-a-time (OFAT) approach feels orderly, but it quietly assumes that each factor acts on the response independently of the others.
When that assumption is wrong, OFAT can point to settings that are not the best available. If the benefit of a longer press time only appears on a prepared surface, testing press time on an unprepared surface will hide the effect completely.
Key Principle
The effect of one factor may depend on the level of another.
What a Factorial Plan Changes
A factorial experiment changes several factors according to a planned pattern so that every level of one factor is observed together with every level of the others. With two factors at two levels, that means four treatment combinations.
Because each combination appears in the plan, the analysis can estimate the effect of each factor on its own and the interaction between them, which asks whether the effect of one factor depends on the level of another.
Efficiency as a Side Benefit
In a balanced factorial plan every run contributes to the estimate of every main effect. OFAT uses each run to learn about one factor only, so it usually needs more runs to reach the same precision.
This efficiency matters, but the main reason to change factors together is logical rather than economic: interactions cannot be detected unless the factor combinations that reveal them are actually tested.
Key Takeaways
- OFAT assumes factors act independently; factorial plans test that assumption.
- Every level of each factor is paired with every level of the others.
- Interactions can only be estimated from combinations that were actually run.