Part seven in a series of blog posts looking at effective design.
In the previous articles we have discussed the design process. Along the way it pays to confirm that the design works by observing your users interacting with it. This ensures the design is as effective as possible before any major development work has been invested, ultimately reducing cost and risk of the project.
Why user test?
You’ve spent hours researching and think you have a good understanding of the problem and user’s behaviour; everyone on the team has tried your prototype and agrees it works well. Why would you want to delay development and conduct user testing?
Simply put, no amount of research can replace good user testing.
Familiarity with your product can distort opinions on the effectiveness of a design; users will nearly always surprise you. They will bring new and interesting problems, ideas, perspectives and highlight flaws you may have otherwise missed.
User testing the new Dynamic Content feature in dotMailer before it goes into development. Here we want to find how easy people find adding dynamic content variations. We record the session so that we can monitor people’s emotions when they use it. We then make changes to the design accordingly.
What are you looking for?
This will depend on the objectives of the design you are testing, perhaps were you aiming for a solution that is easier to use than the current tool or that encourages a certain behaviour. Think about how you can measure the success of your solution, ideally you’ll want a combination of qualitative (descriptive – e.g. user’s opinion) and quantitative (measured – e.g. seconds to complete) feedback.
Who to test?
To get the most out of your testing you will want to test on current actual users, potential customers or at least within the segment that you target. Within these groups it’s a good idea to have a range of skill levels and experience levels. An optimum number of users per group is five. This should ensure the results are balanced and representative of your audience.
Types of user testing
There are many forms of user testing and it can be conducted on anything from paper wireframes, interactive prototypes or work in progress HTML or even a combination of these.
‘Guerrilla’ user testing
This informal technique involves simply taking a laptop to a suitable location (depending on what you are building this could be a conference, a coffee shop or around an office) and getting people to try what you are working on while talking aloud about the experience. The designer carefully observes and makes notes as the user attempts the task.
‘Guerrilla’ testing dotMailer – all you need is a laptop and a notepad.
There are a number of ways to conduct user testing remotely, including using screen sharing software or through specialist services such as TryMyUI that facilitate tests selecting users from their pool of candidates matching your criteria. Similar to guerrilla user testing, user’s actions are recorded as they talk aloud while using your product but you have access to many more users across the world.
An eye tracker is a device that is either embedded in or sits below a computer monitor and uses a number of cameras and infrared technology to record a user’s eye movement over the screen.
This gaze data can provide essential feedback, as users don’t always accurately vocalise their actions, this technique backs qualitative data with quantitative data. The system’s software can be used to generate heatmaps and charts showing how users viewed and interacted with the product.
Testing the template picker in dotMailer. We asked people what they would do if they wanted to create a ‘plain text’ campaign. The ‘hotter’ areas are where most people clicked.
There are pros & cons to each of these methods; you should choose the one that best suits your budget and the stage your project. The important thing is that it becomes part of your process; any form of user testing can highlight issues. We’ve seen that the tiniest change to copy, colour or button position can have a huge impact on task completion levels.
We find that our customers enjoy being part of the process – if you’d like to get involved in user testing dotMailer, become a Test Pilot.