1. Consider statistical significance
Quantitative research aims at reaching an “absolute”. So, you should be taking statistical significance into account. You want to make sure your results are likely to be replicated if the test is conducted more than once.
2. Select the right variables
Design your experiment so that it pinpoints the right variables. The variable you’re putting to the test should be the only thing that changes. For example, if you’re testing a message; with every iteration of the experiment, the message should be the only thing that varies. That means staying consistent on for example visuals, target audience, timing, location, and so on. By adding an additional complexity layer, it’s impossible to tell why a user preferred one iteration over the other.
3. Break down your experiment into smaller pieces
If you’re testing multiple things, make sure you focus on the ones that are priorities, and then proceed to test them one at a time. By doing this, you can iterate your tests based on the results. This will simplify the way you digest the final results.
Now that you’ve got the hang of quantitative concept validation and experiments, we can start by taking a look at what the other kind looks like.
