User testing means giving one or more users access to a website or prototype and observing how they behave when using it. The purpose of this is to discover problems and gain insights that can be used to improve the final product.
The goal of user testing is not to eliminate each and every potential problem on a website – that’s simply not possible (especially if you consider how subjective this can be). The goal is to work towards creating the best possible experience for the user by constantly improving and optimising.
The two biggest questions around testing tend to be ‘What do I test?’ and ‘When do I test it?’ The answers are simple – test as much as possible, as often as possible, and as early as possible.
User testing follows a set process.
1. Formulate a question to test
Spend a little time nailing down exactly why you want to perform a test and what you hope to learn from it. Formulating a simple, clear set of questions to test will allow you to focus on what’s important, and will make choosing participants and techniques easier.
2. Choose a test and prepare
Once you know what the purpose of your test is, you can decide on a specific methodology to use. To choose the right one, answer these questions:
• How much time and money do I have for this test?
• What facilities are available?
• How many participants do I want to test?
• At what stage is the project?
3. Find subjects
Possibly the biggest challenge in the testing process is that of finding the right test subjects. So, how do you do this?
First of all, draw up a list of criteria that you want your subjects to fulfil – must they be men or women, of a certain age, in a certain industry, with or without children?
The considerations can be endless, so limit yourself to the top three or not more than five most important ones.
Now, spread the word about the test through the most appropriate channels to this group. This can involve everything from advertising in a glossy magazine to posting on a Facebook page to chatting to some friends or neighbours. You can also pay a market research recruitment agency to find suitable candidates. The method you choose will depend largely on your budget and timeline, as well as on how many participants you want to recruit.
Once you get enough responses, you will have the chance to screen applicants. Screening is the process of filtering people into those who are suitable for the test and those who are not, because they do not meet certain criteria.
4. Test
At this point, you are ready to begin testing! Tell the user what you want them to do, and let the test run. Don’t interfere!
5. Analyse
Analysing means taking all of this existing data and transforming it into accurate, objective and useful insights.
For example, your user observation study found that users tended to click on ‘contact us’ when looking for the opening times of a restaurant. It’s up to the researcher to analyse this – were the users confused by something? Was there no other obvious place to click? Were they expecting to find this information easily, but found themselves struggling and making a best guess? Discovering the reason
can then lead to possible solutions – possibly the opening hours should be placed on the home page or in the header; or perhaps they should simply be added to the ‘contact us’ page. It’s these practical outcomes that are the cornerstones of UX testing.
6. Report
Reporting is the process of sharing your UX test results with the people who need them. Reports provide insights, information and recommendations by summarising the results of the testing phase, and the UX practitioner’s analysis of what happened. Ideally, the whole team should be involved in analysing the test data to encourage them to buy in to the UX process.
Reporting can take various forms, from verbal discussions to professionally designed presentations. The most important consideration here is your audience and their needs.
7. Implement
Implementing means putting your user testing outcomes into practice. This will, of course, mean very different things at different stages of the project. If you’re testing your overall approach in the beginning planning phase, the implementation could involve taking a new direction on the project. Testing a working high-fidelity prototype may reveal that some design elements need to change.
8. Start again
We’ve said it before and we’ll say it again – testing is not a once-off action, it’s a constant process. Once you’ve run your test and implemented your solutions, your project can continue – but very soon you’ll need to test again. Aim to run a test every time you reach a major new stage of the project, or add something that is brand new or has raised controversy in the team. Even after the project has gone live, there is space and reason to keep testing, iterating and optimising.
The goal of user testing is not to eliminate each and every potential problem on a website – that’s simply not possible (especially if you consider how subjective this can be). The goal is to work towards creating the best possible experience for the user by constantly improving and optimising.
The two biggest questions around testing tend to be ‘What do I test?’ and ‘When do I test it?’ The answers are simple – test as much as possible, as often as possible, and as early as possible.
1. Formulate a question to test
Spend a little time nailing down exactly why you want to perform a test and what you hope to learn from it. Formulating a simple, clear set of questions to test will allow you to focus on what’s important, and will make choosing participants and techniques easier.
2. Choose a test and prepare
Once you know what the purpose of your test is, you can decide on a specific methodology to use. To choose the right one, answer these questions:
• How much time and money do I have for this test?
• What facilities are available?
• How many participants do I want to test?
• At what stage is the project?
3. Find subjects
Possibly the biggest challenge in the testing process is that of finding the right test subjects. So, how do you do this?
First of all, draw up a list of criteria that you want your subjects to fulfil – must they be men or women, of a certain age, in a certain industry, with or without children?
The considerations can be endless, so limit yourself to the top three or not more than five most important ones.
Now, spread the word about the test through the most appropriate channels to this group. This can involve everything from advertising in a glossy magazine to posting on a Facebook page to chatting to some friends or neighbours. You can also pay a market research recruitment agency to find suitable candidates. The method you choose will depend largely on your budget and timeline, as well as on how many participants you want to recruit.
Once you get enough responses, you will have the chance to screen applicants. Screening is the process of filtering people into those who are suitable for the test and those who are not, because they do not meet certain criteria.
4. Test
At this point, you are ready to begin testing! Tell the user what you want them to do, and let the test run. Don’t interfere!
5. Analyse
Analysing means taking all of this existing data and transforming it into accurate, objective and useful insights.
For example, your user observation study found that users tended to click on ‘contact us’ when looking for the opening times of a restaurant. It’s up to the researcher to analyse this – were the users confused by something? Was there no other obvious place to click? Were they expecting to find this information easily, but found themselves struggling and making a best guess? Discovering the reason
can then lead to possible solutions – possibly the opening hours should be placed on the home page or in the header; or perhaps they should simply be added to the ‘contact us’ page. It’s these practical outcomes that are the cornerstones of UX testing.
6. Report
Reporting is the process of sharing your UX test results with the people who need them. Reports provide insights, information and recommendations by summarising the results of the testing phase, and the UX practitioner’s analysis of what happened. Ideally, the whole team should be involved in analysing the test data to encourage them to buy in to the UX process.
Reporting can take various forms, from verbal discussions to professionally designed presentations. The most important consideration here is your audience and their needs.
7. Implement
Implementing means putting your user testing outcomes into practice. This will, of course, mean very different things at different stages of the project. If you’re testing your overall approach in the beginning planning phase, the implementation could involve taking a new direction on the project. Testing a working high-fidelity prototype may reveal that some design elements need to change.
8. Start again
We’ve said it before and we’ll say it again – testing is not a once-off action, it’s a constant process. Once you’ve run your test and implemented your solutions, your project can continue – but very soon you’ll need to test again. Aim to run a test every time you reach a major new stage of the project, or add something that is brand new or has raised controversy in the team. Even after the project has gone live, there is space and reason to keep testing, iterating and optimising.
