See and hear how users experience YOUR website

Find out what they like and what they don't. What's working and what isn't.
Find out WHY they buy and WHY they leave...

How do I get the best results from Usability Testing?

Setting Tasks

The first step to successful usability testing is to decide what you want to learn about.

User Experience testing provides insights that not only help you improve your website, but can also answer many other questions about online user behaviour. Before you start, think about what you want to test and learn.

You can test broad aspects of your website such how easy or difficult it is for customers to find and buy your products or you can target very specific elements such as your search or payment process. You can get some ideas about what you can test here. However, you can test virtually any aspect of your website, applications, customer communications or online media. If you can write a task for it you can test it!

Set a scenario

Once you have decided what to test, a good way to proceed is to think up a scenario. Scenarios help users to put the tasks you set them into context. Here are some examples:


"It's your partner's birthday and you are hunting for a gift"

"Your boss has asked you to research and select an online business card company"

"Your mobile phone contract is up for renewal"


Another technique when setting a scenario is to simulate an email, referral or recommendation.
For example:


"You were in the local bar last night and a friend said that [yourwebsite.com] are running a promotion on BBQs, so you thought you'd check it out"


Sometimes describing a dramatic scenario can set the user's frame of mind - especially if you want to see how fast users can can complete tasks on your website or if you want to simulate how someone in a hurry might use your site. Here are some examples:


"A colleague urgently needs you at a client meeting in another city!"

"Your parents are visiting this weekend and your fridge has finally given up!"

"You are about to leave the house to pick up the kids. You can't be late again!"


Write your tasks

It's important to set a realistic number of tasks per test.

Try to write tasks that can be completed in no more than 20 minutes. Usually, setting one simple task with follow up questions can give you great results. An example might be:


"Use the website to find an outfit for an office party."


If first impressions are important (they usually are) then the first task could be:


"What do you think of the site? What can you do here? Where would you start?"


Also, write your tasks with a clear objective in mind

People normally arrive on a website with a task in mind, so try to reflect that when you write your tasks. Start your tasks with words like:


"You are looking for......"
"You want to...."


And try not to be too prescriptive. You’ll get the best results when users can carry out their tasks in a manner that is natural and relevant to them. For example:


"Find a dress for an evening do that you would genuinely consider buying"

is better than

"Search for a size 10 Black Dress"


Using the word "Find" is better than "Search" because users are more likely to use your website as they normally would. "Search" may divert them to use your search box!

However, there are exceptions !

Sometimes you may want to test a specific part of your site. Maybe your shopping basket or product pages. In this case it's OK to be prescriptive. Here's how:


"1. Find two items you'd genuinely consider buying and add them to your basket. Now go to checkout."
"2. When you get to the basket you've changed your mind and want to add some more items. Go back and find two more items and add them to your basket. Then go to checkout."


Selecting Test Participants

Research has shown that testing with 5 users generally reveals 80% of a website's usability problems. Above that number you don't really add that much more insight. You can find out more about this from Jacob Nielsen, one of the world's leading usability experts.

Most clients test with 5 users, fix the issues, then test with another 5. This is called iterative usability testing and we think it's the best way to improve the User Experience of your site in manageable bite sized chunks.

However, any testing is better than no testing at all and even testing with one user will give you plenty of useful insight and feedback

It is better to test often with a small number of participants than to test once or twice with many.