Null hypothesis


When conducting tests using statistics, a null hypothesis is used in conjunction with a hypothesis in order to determine if there is statistical significance that warrants proceeding with an experiment.

A null hypothesis states that there is no significant difference between a set of given observations. In the case of a marketing experiment, this means that a null hypothesis is proven true means that there is no significant difference between changing the variables used in the test, and not changing them.

An example would be an A/B test on marketing copy in an advertisement.

The hypothesis would be that changing the marketing copy to variation “B” would have a better outcome than using the copy in variation “A”.

The null hypothesis would state that there is no statistical significance between the two (A or B).


Statistical significance

Multivariate testing