Another way of discerning different kinds of tests is by considering whether a test is done in a ‘lab’ or in the ‘field’. In lab tests, tests are conducted at a  special location where conditions can be controlled. For example, for functional testing, products are put into a climate room. Usability tests are conducted in what is called a usability lab, filled with cameras, where test persons perform tasks designed by the researchers. An example of a market test in a lab is when consumers are invited to taste and evaluate new products.

The advantage of lab testing is that the outcomes are reliable: when the same test is done later in time or with other researchers it is likely that similar results will be obtained. The downside, however, is the lack of relevance. Without context, there is a risk that the lab test isn’t replicating the real world, just mimicking at best. This is a huge risk with user and market testing. If you let test persons perform specific tasks in a lab, the question is whether people would do these tasks in the real world at all, or whether they would do it in the fashion imagined by the researcher.

Field tests are tests performed in the real world, offering real-world context. Functional tests in the field are e.g. prototypes of new car models that are driving around thousands of miles (often camouflaged). User testing in the field are, for example, beta tests, which we will discuss later. Market tests are mostly conducted in the field, as e.g. products are sold at a small scale in a limited area to learn how it performs. 

Field test offer relevance: they do not mimic the real world, they are the real world. However, outcomes are often less reliable, as small changes in the set-up or conditions of the test, may already result in different outcomes. For example, when a new product is distributed in a few shops and sells well, the success may result from the fact that some shop owners deliberately pushed it, as the test gave them a competitive edge compared to other shops. Therefore, interpreting results should always be done carefully.