Usability Testing, The Main Techniques

A Brief Introduction

Hi all, I’m Matt T, a designer at LCN.com. I only joined the company in June so this is my first blog post. I have a keen interest in the Usability & User Experience aspect of web design so thought I’d write about that. I hope you find it useful and would love any comments you have. Anyway, let’s get on with the show…

An Overview of Usability Testing

Usability testing makes sure that your website can be used effectively by your intended audience. It helps reduce the friction and annoyance people experience when things aren’t as easy as they should be. From a site owner’s perspective, it’ll make your website more effective and more popular.

Why you should test your site with real people.

One of the biggest mistakes people make when designing a website is that they assume that their users will have a certain level of knowledge, and think about things in the same way as they do – and they’re often wrong.

Problems like this are hard to pre-empt without spending a lot of time on user research before starting your project. It’s also extremely difficult to look at your own site with fresh eyes and to try and think in the same way as your users. That’s where Usability Testing comes in.

The Main Types of User Testing

Usability Tests & User Research fall in to two main categories – based on the type of results they produce. These are quantitative and qualitative tests.

Quantitative User Tests

Designed to be highly measurable, Quantitative tests provide statistics that prove or disprove an idea or theory. They tend to be run with larger numbers of participants – often on a live site. They include:

Time-to-Task

Measuring how long it takes a user to achieve a certain tasks, or goal – for example how long it takes to find, or purchase a product from an e-comerce site.

A/B or Split Testing

This is where you divide the people visiting your site into groups (usually two) – and show each group a slightly different version of a page. You’d then record the actions a user takes after seeing that button.

When you review the results you’ll be able to see which variation produces the best result.

An example would be showing two different versions of a “Checkout” button – and seeing which results in the highest number of orders.

Qualitative User Tests

Qualitative tests are intended to provide insight, rather than to prove or disprove a specific idea. They’re often run with smaller groups – often only 4-5 people – and often won’t provide statistically significant results.

However, their great advantage is that they will often give you answers to questions you haven’t even asked yet. Qualitative tests can provide real insight and understanding of how people interact, and think about, not just your website – but your business and industry as a whole.

The Five Second Test

The five-second test is used to test a specific web-page rather than an entire site. It’s designed to make sure that your page is communicating it’s most important information properly. It’s also one of the simplest (and quickest – as you’ve probably guessed from the title) tests you can run.

Simply show someone the web-page you’re testing for five seconds – and then hide it. After that – simply ask you test-subject what they remember about the page, what was important and what actions they can take on that page.

You’ll quickly discover if the information you thought was obvious really is. Why not test your own site at fivesecondtest.com.

Observational Usability Tests

This type of test simply involves watching somebody using your site as they try to carry-out a pre-assigned task. These tests are often recorded using software such as Silverback by Clearleft

The user is asked to ‘think-aloud’ – to give you a better insight into what they’re thinking as they move around your site. You have to be very careful not to help the user – or ask leading questions that will influence their behavior.

You’ll see where they get stuck or confused – not only by looking at their screen – but also their expression and emotional reactions.

You should be able to catch most of the usability issues with your site with around 4-5 testers.

One Response to “Usability Testing, The Main Techniques”

George - Planet Anarky Says:

October 13th, 2009 at 11:15 am

This is a good post that gives some great starter points on usability and how to test whether a design is effective or not.

Usability is quite often overlooked, as you say: it’s too easy to get cabin-fever with a website look & feel and end up forgetting the basics of how people want to interact with a site.

Personally, I feel the Financial Services industry is one of the biggest culprits in this arena.