Different products have different requirements, so it’s very important to customize each and every step to serve your goals better.
Create a task list
The number one step is to Create a list of task scenarios you wish to test. Once you have a list prioritize them based on their importance. Every application/ interface/software has certain critical paths which can make or break an experience.
They are also called the red routes and will capture 90% or more of your user’s actions.
On a travel booking platform, booking a room in a hotel or booking a file is a red route that must be fixed first before anything else.
Create a checklist of heuristics
The second step after identifying the critical paths is to Finalize and create a checklist for the Heuristics you are going to use. The Nielsen’s Heuristics are of course a good point to start, but are there any other heuristics that are important? Can you customize them further depending on the domain?
At this point, you must also have the brand guidelines or the design system handy.
The third step after identifying the critical paths and finalizing the heuristics is to appoint the evaluators. Who are the people who are going to audit the UX?
The ideal number is between 3-5. Anything less than 3 will be more of an opinion and not an honest review.
Regarding the qualifications of the evaluators, the ideal scenario would be to hire both 3rd part UX experts and domain experts to have non-biased opinions.
But since that may not always be possible, try to appoint at least 1 UX expert and evaluators from teams other than the team that has directly worked on the product.
If you are working with a client, including people from their side can actually be a good strategy as the feedback may not be vague and may be based on a solid and tangible framework.
Also, you need to appoint a reporter. The job of this person will be to collect the evaluation from each evaluator and find common and most critical issues.
Ideally, this person should be someone who is a UX or Usability expert but did not participate in the evaluation.
This is to ensure that there is no bias in the reporting of issues.
Now that the roles are defined, the fourth step is to hold a briefing session for the team, where the heuristics, the red routes and the timelines etc are discussed to put everyone on the same page. At this time it is important the team also establishes the common method of reporting severity.
For example – Severity rating one is Most critical issues or is it the light issues.
Do the Audit
The fifth step is to actually conduct usability audits with heuristic (evaluation). Evaluators need to do it separately and not together.
At UX Gorilla, we encourage evaluators to also include their recommendations or any references for the interfaces that have implemented the same functionality in a better way without violating the heuristic. The individual reports are then sent to the reporter. The job of the reporter is then to comb through the reports and find common themes and prepare a report with the recommendations. This report is then presented to the stakeholders and the next sprint is planned.
This process has worked great for us but it has evolved a lot over time. You are free to tweak it as you like. The goal must be however to do this often and early. So do not wait for the product to go live before you do the evaluation. Do it even if right now you only have a prototype.