Full Factorial

Definition:

Full factorial is a multivariate design of experiment (DOE) where all possible combinations of variables (where a combination is called a recipe) are included in the test.

For example a test with 3 page elements (e.g. heading, image and asset) with 3 levels for each, would have 27 combinations or recipes (3x3x3). For this reason a full-factorial design takes much longer than a fractional-factorial design to complete, but it has a much higher degree of accuracy and is less likely to lead to incorrect conclusions being drawn.

Comparison of Full Factorial and Fractional Factorial Design

Image of table of comparison between full factorial and fractional factorial design

Source: Gupta 2011

Resources:

Conversion marketing – Glossary of Conversion Marketing.

Over 300 tools reviewed – Digital Marketing Toolbox.

A/B testing software – Which A/B testing tools should you choose?