The most common mistakes when planning a usability study

The most common mistakes when planning a usability study

Usability Testing has become an intrinsic part of software development over the last decade. Virtually any software development—be it house, web development, or design—depends on usability testing to validate and prove their hypothesis in user interface design. Production, marketing, and QA aid greatly in such an endeavour and help the organization better orient its products and offerings.

Innovation in the field of biotechnology—specifically in brain-computer interfaces, neurotechnology, and biometrics—have enabled organizations to validate their findings and hypothesis based on real-time bio data analysis. However, these advancements come at the cost, and poor experiment design risks failing the study.
First, and as in any market research, one should consider the demographics. I will spare you the conventional wisdom about demographics as many have established well proven theories about it. However, when it comes to Neuroergonomics and usability testing, what we are analyzing are biological data, and this is related to the person’s cognitive functions. Hence, designing a study should take into consideration the mental condition and cognitive functions of the subjects being tested especially in relation to their age. For instance, you can't give a 10 year old a CRM system to test and vice versa you can solely rely on a 40 years old to test a newly launched video game or social media platform. In addition, it’s important to consider the availability of your test subjects: testing pilots for example would require them to be available for the study, and in the case of engineers and other regulated professions the real users may be scarce.

Talking about availability, one must think about the availability of the environment. While this can be easily mitigated nowadays by using simulated or virtual reality environment, the real world is completely unbiased! Of note, clearance issues and scheduling complexities start to arise especially when running into high security places like manufacturing, military and so on.

Planning a usability study as in a simple A/B testing procedure may result in a lot of overhead cost. Not to mention the release of an “A” interface and its failure to prove the hypothesis and all the cost incurred running it live for a certain period of time and loosing sales in an e-commerce scenario. (Neuroergonomics bring a solution here by testing it on a subject group) but the cost arises and if the tests fails, well we only lose the negligible amount invested in the study and not the loss in sales and overheads.

By minimizing all the risk in data collection and study preparation we narrow our losses, generate a more meaningful outcome from the study, and make sure we successfully answer the initial hypothesis. Despite the impact of an unsatisfactory result, it is crucial to be unbiased and not bend the information, or otherwise risk losing the confidence of the stakeholders. 
It is important to be unbiased and not bend the information for a It can be a negative or a positive result however the option here is to prove a hypothesis and not to bend the information to always get a positive result. In the case of Negative results, we face the problem of impact and the effect the study leaves on the Owners, specially if the results do not confer with their opinion, hence if not planned well and well explained from the beginning you might easily lose their confidence.