Question: Which of the following is a best practice when implementing an experiment?
- Start with multiple hypotheses.
- Assign two campaigns to each experiment arm.
- Test one variable for each experiment.
- Divide your audience into as many arms as possible.
Explanation
A valid experiment needs a clear control and treatment so performance differences can be attributed to the change being measured. When only one campaign element differs, results are easier to interpret and act on. Multiple simultaneous changes make it unclear which factor influenced performance. This structure supports reliable creative learning and cleaner optimization decisions.
Why the other options are incorrect
Multiple hypotheses: Several hypotheses create overlapping variables and reduce result clarity.
Two campaigns: Campaign count alone does not ensure a clean experiment design.
Many arms: Too many arms can fragment traffic and make results harder to read.
Source for verification
https://support.google.com/google-ads/answer/13719071
The answer(s) to the question is highlighted in the BOLD text above. You can also find more questions and answers related to the exams on the "Google Ads Creative" page.