# Usability Testing

Peak Shift uses usability testing to uncover problems and opportunities in a client's application design. We also use this popular observational methodology when gathering data on how applications and their various user journeys are performing over time, identifying which areas require design attention in upcoming sprints.

In a typical usability-testing session, a researcher asks the participant to perform certain tasks, usually interacting with a specific interface(s) or user journey/flow. While the participant completes these tasks, the researcher observes their actions, behaviours, successes and failures, as well as listening to any feedback or anecdotes they have along the way.


**The goals of a usability test are:**

- **Identifying** any problems or painpoints in the application's design
- **Exposing** any opportunities for improvement
- **Observing** and **learning** about the user’s behaviors, patterns, and preferences
- **Avoiding** blind design bias from designers, marketers, and product owners

## How to prepare

**Step 1:** Identify the application feature, journey, or service you wish to test and analyse. Create a clearly defined task for the user that can be tested. When designing new features (as opposed to analysing existing ones), this often involves creating a working prototype, otherwise you can just use your current application or platform for testing.

**Step 2:** Set a timeline. Usability testing can run for a day, week, month, or even a year. It depends on how much data you wish to collect and how many participants you choose to analyse. (We recommened somewhere between a day - week).

**Step 3:** Find your participants. Again, depending on how much data you wish to collect you will need a certain number of participants to take part in your study. In order to avoid wasted efforts, you should perform an initial screening test to make sure you are asking the right questions to the right people. Make sure to use realistic users of your platform, or users of a similar platform. It is good to keep the participant pool diverse to avoid any bias (this includes power/new users, as well as ethnic, cultural and socio-political background).

**Step 4:** Prepare your test. This is usually defining a set task(s) (e.g "Go to www.bank.com and set up a Current Account"). If you are testing a new feature, make sure your prototype is ready to go and users are clicking random links for 10 minutes wasting unecessary time. Make sure to note down any follow up questions you wish to ask the participant.

You should also decide whether your usability test will be ***qualititive*** or ***quantitive***. Qualititive usability testing focuses on observing a user and collecting insights, observations, and anecdotes. This type of testing is useful in discovering UX issues and pain points for your users, as well as collecting feedback on how to improve your application. Quantitive usability testing focuses on collecting certain metrics and data that help measure and analyse the user experience. The two most commonly collected metrics are task success rate (did they complete the given task?) and time on task (how long it took them).

**Step 5:** Reward the participant. It's important to keep your testers happy, as well as being polite and professional. Gift vouchers, in-app experiences, and other small thank-you's can go a long way.

## How it works

The researcher guides the participant through the test process, and provides the participant with a specific journey or task(s). As the user performs the tasks at hand, the researcher observes their actions, behaviours, and frustrations, as well as listening for any feedback along the way as they are asked to "think out loud". The researcher's role is to keep a written track along the way, data which will later be analysed. Either during or after, the facilitator can also ask followup or prompting questions to extract additional detail from the participant.

In order to ensure that the test results are high-quality, accurate, and valid, the researcher must make sure they remain neutral throughout the testing process - avoiding accidentally influencing the users behaviour or guiding their actions.

The researcher collects the results of the test (whether qualititive or quantitive), and can then cross analyse it with that of other users. This was researchers can spot user behaviour patterns, frustrations, and successes, as well as noting anomalies or inconsistent results to avoid mishaping a product's design direction. Tests should also be recorded (if the user gives their permission to do so) to avoid losing any information along the way, ensuring all relevant feedback has been recorded.