How to Utilize a User Testing Platform

When clients have highly specialized users or our internal teams need help understanding how and why users interact with a flow, I conduct a usability test. My experience comes from utilizing Lyssna through my previous agency so I will show my process through that platform.

For a more in-depth overview of usability testing, check out my post on it here!

Overview

Through Lyssna, I can recruit participants who match a client’s target audience and perform moderated and unmoderated research. 

While Lyssna is my platform of choice, many others can be utilized to do a variety of user-moderated testing, similar to Lyssna. These platforms provide a variety of testing options depending on a client’s unique needs, including:

Design Surveys – Offer feedback about a design – whether a website, app design, animation, prototype recording, video ad, or logo concept. Design surveys display website or landing page designs and questions for participants simultaneously, allowing respondents to inspect the design closely while moderators observe.

Website Testing – Whether manual or automated, website testing ensures a bug-free, user-friendly, and efficient web experience.

Prototype Testing involves creating and testing a prototype with real users to validate design decisions. 

Five-Second Tests – A five-second test is run by showing an image to a participant for just five seconds, after which the participant answers questions based on their memory and impression of the design.

Card Sorting – Card sorting is a well-established user experience (UX) research method where participants group topics into categories according to criteria that make the most sense to them. 

Preference Testing – When taking part in a preference test, a participant is shown several design options and asked to choose one. We often use these tests to measure aesthetic appeal and to judge designs based on their trustworthiness or how well they communicate a specific message or idea.

First Click Testing – First click testing aims to verify that a user’s first click on an interface is straightforward when asked to carry out a given task. 

Navigation Testing – Navigation tests analyze how users navigate a website or application when presented with a specific task or goal. The results help us understand critical user flows and improve information architecture.

Tree Testing – A tree test evaluates a website’s hierarchical category information architecture structure. Tree testing asks users to find the locations in the tree where specific resources or features can be found. This is helpful for testing site migrations or when adding new pages to a website to ensure the website’s content organization, structure, and navigation are intuitive and user-friendly. 

Planning

It is vital to set clear goals for the usability test to help clarify my process and the research questions the study will attempt to answer. The usability test plan should outline the tasks I will ask the participants to accomplish and all pre-task and post-task questions. This will all be reviewed and approved by the client before I begin recruiting for the test.

Set-Up  

Once logged into the platform, I am immediately taken to the dashboard where I will create a new project for my client (you will see our existing client projects in the screenshot below) and then click “create study” within the project.

Above Image: Illustrates dashboard view of lyssna.com

For illustrative purposes only, I will use an example of a user survey we performed for our B2B client in the restaurant industry. 

Step 1 – Test Details

After I create a study I am brought into the “test details” dashboard where I will implement my study information. In the image below, l have listed my test name (it is important to differentiate your naming structure when performing multiple tests on different devices) and which device I will test on. 

Step 2 – Setting up the welcome screen

The welcome screen illustrated in the image below will be the first page users will see when they are selected to complete the survey. It is on this page where I will set the scene to give the survey participants some context on the questions they will be asked to answer.

Above Image: The welcome screen as seen on the backend of the Lyssna platform

Above Image: The welcome screen as seen by participants when taking your survey on the front end.

Step 3 – Instructions & Setting Expectations

The third step is to add any specific directions or insights I wish to provide participants with, as I will have no other opportunity to explain the nuances for which information I hope to gain from their answers once the survey begins. It is essential to use clear language that does not push the survey participants in a specific direction or create bias that may cause them to answer my questions in a way that will undermine the survey outcome.

Above Image: Adding instructions for the survey participants within the Lyssna platform.

Above Image: The survey instructions when presented to survey participants on the front end.

Step 4 – Design your survey 

My next step is to choose the specific survey/test. Each test provides specific instructions for the video, audio, images (including size limitations), and content you can implement. I can customize directions for follow-up questions dynamically based on how participants answer.

Above Image: Shows the test options available for use with Portent’s membership within the Lyssna platform.

Recruiting

When my test has been built with I’s dotted and T’s crossed, then the final step is to recruit users to participate.

As illustrated in the image below I can recruit from either my client’s user base or recruit anonymous users from Lyssna’s panel of experts.

If I choose to recruit from my client’s user base, I will want to schedule this with the client as they can provide an incentive to participants to complete the survey. While this process often takes a bit longer, it provides personalized data that can be used for analysis.

Often I will want input from users who may not have firsthand knowledge of my client’s business to get a fresh perspective on how I can make meaningful changes to a website experience. In these cases, I will recruit from Lyssna’s panel.

Above Image: Two recruitment options within Lyssna’s platform.

When recruiting participants from the knowledge panel, there are many ways I can segment our audience to get the most qualified answers. The photo below illustrates how I can segment the audience by country, age, income, etc. There are several more ways I can segment by Industry and occupation, Financial products (home ownership status, if they have investment portfolios), as well as by technology (proficiency, mobile vs desktop, if they use food delivery or mobile dating services), and hobbies/interests.

The cost for each survey will depend on how many participants are recruited and in which country they are recruited in. Often I use a small sample size of about 20 – 50 users to keep costs low and still gain valuable insights.

Note: The more in-depth you are with segmenting users, the more time it will take for Lyssna to find qualified participants which may slow down turnaround time. However, normally responses will filter in within 24-48 hours after purchase and be completed within 2 days.

Results

Once Lyssna collects the responses from participants I can see their full responses within the platform.

I am also able to see general demographic data on the participants within the sidebar which helps me see if users skewed a certain way because of their demographics.

Note: At times you may find with smaller sample sizes that an audience may be skewed. For example, because participants are chosen at random there may be a time where I will receive more male participants than female and the men gave much different answers. In this case, a re-run of the survey may be warranted to correct for any biases.

Lyssna is very good at visualizing participants’ responses in succinct and easy-to-understand tables and graphs – I will often use these in my presentation decks to represent audience responses.

Above Image: An example of participant’s responses to a ranking question within the Lyssna platform

I will download all responses as a CSV. This allows me to organize it so I can highlight patterns and trends that can make the most impact in my presentation. 

The trends identified should center around elements that create friction for the user. These can be elements that did not meet user expectations, needed to be more transparent for users, or acted as a conversion barrier in some way. It is also helpful to identify trends in unanswered user questions and user priorities when seeking information.

Note: Though friction points are the best source of optimization ideas, points of delight can also be impactful to focus A/B tests on. I’ll look for content and on-page elements that users respond well to, and communicate that to my client.

Presenting Your Findings

I share my usability test results alongside CRO roadmap recommendations when possible. This is often an hour-long meeting with the client where I will explain my survey findings and suggest any recommendations that I can implement into my testing roadmap.