This week I discovered some clarity on the distinct types of usability testing. For years I have been performing different methods of usability testing like 1:1 lab sessions, heuristic evaluations, contextual inquiries, focus groups, surveys and AB tests to produce results. Over the years, I typically refer to this diagram from Jakob Neilsen to assist what user research method is appropriate. However suppose I did not think that these methods may live under different types of usability studies. I am now clear on what the main types of usability do. As a matter of fact, I found this image to refer to what usability type to pick and I did a quick comparison as well below. I think having a clear understanding of each usability study type will help user experience professionals better use each research method.
The two main types of usability studies are formative and summative. A formative usability study is typically performed at the beginning of a development cycle. On the other hand, a summative test is typically an assessment generally at the end of a development cycle before a product launch. A formative study is testing with representative users and representative tasks on a representative product where testing is designed to guide the improvement of future iterations. A formative test is exploratory and determines what is not usable. A formative test is used to diagnose rather than evaluate. While a summative study is how usable is it. The goal of this test is to evaluate rather than diagnose. A formative study will undercover 80% of discoveries with a small number of participants in a short amount of time. A summative study requires a larger number of participants to get reliable data and can be time consuming and expensive. A formative study provides support to enhance the product since the focus is finding usability flaws A summative study measures effectiveness, efficiency and user satisfaction.
So do you have anything to add? Is this a clear breakdown? I hope this inspires you on your next usability study.
