UX as a science: Your key to heuristics
Analysing an analysis
In UX design, we perform a heuristic analysis as a quick and inexpensive way of testing the usability of an interface.
Don’t get me wrong. In no way does this replace user testing but can serve as an FYI to what improvements can be made. I’ll always emphasise the need for user testing.
In this article, however, we’ll focus on what constitutes a heuristic analysis and how to set one up of your own.
Understanding the basics: What is a heuristic analysis?
As a broader term, heuristics can refer to a learning or discovery process that occurs through trial and error. (Formed from the Greek word heuriskein, meaning, ‘to find’.)
Heuristic Analysis/Evaluation/Review is a usability engineering method for finding the usability problems in a user interface design[…]. Heuristic evaluation involves having a small set of evaluators examine the interface and judge its compliance with recognised usability principles (the “heuristics”).
— Nielsen Norman Group.
When Nielsen manifested his ten heuristics for usability, he intended them to be utilised in a broader context so that they can be applied in various domains. Outside of UX design they’ve been applied to hospital layouts, educational material, and much more!
However, Nielsen specified that these heuristics need to be modified and made more contextually specific for a more accurate evaluation.
A key; but for the wrong lock
A while back, I was tasked with performing one of these analyses, which seemed straightforward. At least, that’s what I thought …
The checklist provided was intended for web evaluation (traditionally on a desktop display). It included verbs such as “click” and interface elements such as “page” and “window”. After doing some research, I discovered that UI heuristic checklists have until recently been directed towards desktop evaluation; and people still use these same checklists for mobile?! It’s like having a key, but for the wrong lock.
To strengthen my case, imagine evaluating the usability of virtual reality (VR) using a mobile checklist… there’s no way people will take your analysis seriously! Why should mobile vs desktop be evaluated to the same criteria?
What makes mobile different from web?
As we know, mobile has unique constraints that determine how we will design for that specific interface. These constraints differ from a laptop, desktop or smartwatch screen. They are distinguished by the following:
• Limited input/output facilities
• Mobility and varying context (quick access)
• Types of tasks performed
• Multi-device access (interconnectivity)
• Limited processing capabilities and power
• Adoption (what makes us adopt certain tasks on mobile over other devices)
The study of mobile constraints is necessary to identify mobile-specific heuristics. Rosa Yáñez Gómez et al. have collated a list of such heuristics that can be downloaded here.
Step 1: Always be aware of your own bias
“By their very nature, heuristic shortcuts will produce biases.”
- Daniel Kahneman, Nobel Prize-winning economist
As with all qualitative analyses, there is always an element of bias.
When evaluating an interface, designers need to be aware of their own biases or opinions they have that steer away from the checklist in the slightest — those are inherently subjective. When performing the review, strictly follow your checklist, taking your time to avoid taking shortcuts or leaning on your personal opinions.
Heads-up: A heuristic analysis, especially done well, takes time. Cater for one to three days if you review a short to medium screen flow.
What if you don’t have the luxury of time?
A checklist needs to be as extensive as possible. It brings an evaluation as close to being scientifically accurate as possible by avoiding assumptions.
However, if you are short on time, I recommend that your first priority is covering your bases on Accessibility. WCAG and W3C are good resources for this purpose.
Usability of the checklist itself
Your checklist is for the evaluator and the stakeholder and should thus be suited to both.
Your evaluator might not be a UX designer. They could be a technical lead! For this reason, it needs to be as detailed as possible to guide whoever will be using it.
Evaluate only short flows at a time so as not to exhaust the evaluator. An exhausted evaluator takes shortcuts!
The analysis will eventually be presented to a business lead, meaning it needs to be easily understood. Even though the checklist is long, the summary and process should be simple to explain to stakeholders.
To summarise, the checklist needs to be as detailed and all-encompassing as possible and, at the same time, be presented in a simple way that makes it easy to understand within moments.
Building your ultimate checklist
I would recommend any UX practitioner, who has the time, to collate their own or start with an existing list, such as Gómez’s, and refine its usability.
Here are a few guidelines for you to build your own:
- Ensure that the heuristics at hand are relevant and that they follow well-recognised principles. Comparing sources is your friend. Accessibility is always a good starting point.
- Ensure that the checklist is easy to read and understand.
- Every word must pull its weight, especially for a lengthy analysis. Conciseness is key!
- Categorising or theming on multiple levels is recommended.
- Gómez et al. suggest you have main heuristics (such as Nielsen’s heuristics) and sub-heuristics (more elaborate points below the main heuristics). This makes it easy to present scores for overarching heuristics from a high-level view.
- Customise the checklist. Your checklist will change. It’s OK to omit criteria if it isn’t essential or when operating systems update their guidelines and new patterns are adopted by users over generations.
Conclusion
In this article, you will have learnt what a heuristic review is, how to compile one of your own that is suitable to your context and how to use it yourself. Mine is continually under construction, and so should yours be!
Sources consulted:
https://www.researchgate.net/publication/299460619_Establishing_usability_heuristics_for_heuristics_evaluation_in_a_specific_domain_Is_there_a_consensus
