Before designing your test you have to be aware of following:
Test persons will never answer you correctly.
- Some persons will answer socially correct in the answers but not in reality. Some are overcoming this problem by asking what the test person thinks the neighbour would do.
- If you are asking on a question in another situation you will have another answer. If you ask about the shopping for the dinner after the dinner or in the morning you will not have the stress the consumer have ˝-1 hour before dinnertime.
Make the interviews as near in both time and place to the real purchasing situation. The best position will be in the supermarket as near to the product department as possible and as late afternoon as possible for daily-products.
Test persons do not have time for very long interviews. You must limit your questions to interviews at about 10-15 min.
Another important matter is the number of completed interviews you need to have statistical sufficient data. A general rule will be to have min. 150 completed interviews plus extra 50 for each new design you are adding over 3 designs. In order to have 150 completed interviews you need to stop about 300 persons and use about ˝ hour at each completed interview.