Archipelago

You have a website, and you want to make it work harder for you. What should you do?

One approach is to think about extra features you could add to the site, such as forums, client login areas or social-networking tools. You could try to figure out what changes to the site will improve your lead generation or sales conversions.

Trying to predict what users might want from your website (and where they might get stuck) is fine but, ultimately, you’ll just be taking a guess. Which is all well and good, as long as your guess is right!

If you make changes to your site based on guesswork, it could cost you thousands of pounds and months of head-scratching to find out you were wrong. Even worse, you might never know for certain why your grand redesign led to a drop in enquiries.

The bottom line is that you and I don’t know – but we know who does – your website’s users.

So how do we find out?

‘User Testing’ is where real people are closely observed using your website. They are assigned a set of tasks (finding a specific piece of info, or buying a particular item) and then watched to see exactly how they interact with the site.

This gives first-hand insight into how people actually use the site, and allows the identification of any obstacles that stand between your visitors and their goals.

The first step in the user testing process is the ’5-Second Test’. The user is shown a particular page (your home page, or a key product page) – for just 5 seconds and then asked to describe your company, and the product or service. This gives a very clear idea about what your site is (and isn’t) communicating effectively.

After this comes the Key Task Testing. The person’s mouse movements are tracked, and they are filmed as they try to accomplish various tasks on the site. They are asked at key stages what they are thinking, what they’re about to do next and why.

By videoing the user’s face you can see their emotional responses as they use the site.

This process helps build a reliable picture of how people are using the site.

Once the user testing has been completed, the site should be reviewed using a simple ‘triage’ process to diagnose and prioritise the big problems and the easy fixes.

This is the best way to make informed decisions about what you need to fix, and in what order. This reduces the risk of spending money on changes you don’t need.

Is it appropriate for any site?

Whilst high-end testing for sites such as Amazon is done in a lab (with eye-tracking systems to see where someone is looking on a web page) and requires a high-end budget, an experienced web design agency can get most of the benefits of this process simply by watching four or five people use a site to complete a set of tasks. For many sites this should amount to just two or three day’s of an agency’s time, depending on the number of tasks defined.

How much of a difference can it make?

Even the most basic user testing can show up massive differences before and after site changes – such as the average length of time taken to find a particular piece of information. We recently re-designed a website where we reduced the amount of time taken to access key information from from 30 to 7 seconds.

The practical impact of this change was to reduce the websites’ bounce rate (the percentage of users leaving site after only viewing one page) by half.

There’s a lot of anecdotal evidence on savings made and increased profits achieved through user testing. For a great example see The $300 Million Button – where simply renaming a button and changing a few words increased a company’s revenue by 45%!

Whatever the size and scale of your website, you should certainly consider user testing as a way to get the best performance out of your investment.