As far as the idea, or product, you would like to test, it should be something that is interesting to you. The whole purpose of doing a scientific experiment is developed from the standpoint of a question you asked about something, where you thought there might be a relationship between one factor and another. For example, one might wonder if there is a relationship between the amount of sunlight a plant receives and how much that plant grows. One would design an experiment around that question that would either confirm the fact that plants do better with more sunlight, or dispels that fact by showing sunlight is bad for plant growth.
The experimental group in any experiment is the factor, called an independent variable, that you think makes the difference. It is the factor you change and then measure the outcome, which is called the dependent variable, because it depends on how much of the independent variable was administered. The control group in any experiment is the group where you keep all the variables the same, you don't change anything. The control group is the "norm", the measuring stick against which you measure the progress of the experimental group.