2004

Gary Flood

First published in Internet World

So your new website is fast, slick, got loads of content, and you can buy all your company's goods online off it over a highly secure link. Time to sit back and count the millions as customers stream to your virtual shop front, right?

Alas, you may have to postpone the champagne cork popping just a little longer, because the most vital factor in whether a website is actually used may have been overlooked. How sure are you that your shiny new e-commerce zone is actually usable by the people you want to buy from it? Unless you can answer 'because we asked them,' you may be in trouble, according to a growing chorus of experts (and converts) who say that usability is still a major downfall of UK.com.

For example, in June, design consultancy Salmon delivered a system for Argos that can find any product from the firm's entire 12,000-item catalogue in three clicks.

The fact that Argos - which claims to be one of the three most popular websites in the UK - felt the need to do this should tell you something about the state of usability. The best websites are the ones that don't just put up the best-looking template their web design agency came up with. Instead, they either start from scratch with a so-called user-centred design approach, or only work with those designers who understand wonderful new things called information architecture. Or they take the time and trouble to actually field-test their web shops with real punters, who may find your stellar design doesn't actually help them to do what they clicked on your site to do - research or buy a product.

Don't believe you need to worry? Dutch e-commerce consultancy Fredhopper this year analysed 500,000 consumer website visits. It found less than 10% of visitors to retail websites make a purchase, a worryingly high 20% to 30% of potential customers explore no further than the home page, and 65% of visitors leave after viewing only one or two pages. So why are we bothering?

Similarly, UK start-up website performance evaluation specialist SciVisum claims its equivalent research found 81% of large UK companies rely on customer complaints to improve website services. (Think of how comfortable you'd be with that as your main research process for product improvement.) Put this and other such data together and you soon start to see a disturbing pattern: the missing link in far too many websites may be the customer you're trying to reach and retain.

'If you haven't done user testing, frankly I don't think you should be on the web selling things,' is the take-no-prisoners viewpoint of one of the undisputed pioneers of usability thinking, Jakob Nielsen, principal of the Nielsen Norman Group.

Nielsen - and many others - say they're not banging on about usability just for pure aesthetic reasons. Poor design equals poor sales. If design is bad, it will put users off too easily, or it won't allow them to navigate or cruise the site the way they want to, and that feels most natural to them; it will put too many obstacles in the way of the main purpose and make getting information or closing a purchase too klunky. All that could be robbing you of sales - and sales targets - that you never knew you ever had.

Let's put that in a more business-oriented way. Quickest way to double the amount of money you get from your website? You can either increase the number of visitors by 100% - or double the amount they spend once they're there.

Guess which is easier and cheaper? The latter will cost around 10% of the investment the former would, according to Nielsen and other usability doyens, such as Paul Blunden, chief executive of The Usability Company, a small consultancy based in Clerkenwell that's been talking to companies about how to improve website usability since 2001. A case in point is Danish online retailer Bon A'parte, which carried out a big usability research project at the end of 2002. 'This told us a lot of things we didn't know we were doing wrong,' says its e-commerce manager Brian Andersen. Immediate changes made were fixing the index page, allowing more browsing of the catalogue, and other navigation changes. The result? 'We got an immediate improvement in conversion rates that quickly paid for the work.'

Another example is financial services firm Charcol, which used Blunden's company to help find ways to improve conversion rates on its www.charcolonline.co.uk site. 'We found some of the results quite horrifying,' says its head of e-commerce Russell Gould. It turned out the system wasn't flexible enough to handle the many ways people (naturally) write money figures out, like £100,000, 100K, or 100,000.00, and date formats were also different in different parts of the site. 'Customers were getting kicked back a lot by things we wouldn't have seen without testing,' he adds. But a few quick changes resulted in happier customers, a drop in calls to the call centre, and almost immediate hikes in conversion rates. But surely this should be the role of the designers? Good website layout and functionality is what we pay them for, surely?

The reality is that until recently too many designers concentrated on the 'wow' factor, and may have lacked the in-house usability expertise they should have. As Nielsen puts it, 'That spinning 3D view of the product your web design agency says is cool should only be there after you have tested that real users can view it without being confused.' Blunden adds: 'You can't assume your web design agency does this - you should talk about it from the start and make sure they can demonstrate they have, or can access, the expertise.' If nothing else, many wheels may have been reinvented. 'I'm amazed at how many companies don't just steal the ideas that obviously work from sites like Google and Amazon,' muses SciVisum's chief executive Deri Jones. Leading internet marketing technology companies like Agency.com agree. 'In the UK, we think of usability as a bit fluffy,' says its head of information architecture, Clare Munday. 'That means clients don't want to pay for it; so if your design agency doesn't do it inherently, it may not get done at the right stage, from the start. US and Asia-Pacific clients are much more demanding on this as the best customer experience is a high value for them.'

Luckily, there are a number of ways sound design principles can be applied to either guide design of a revamped website, or lead from the start. But many companies won't have the luxury of a complete overhaul. The good news is that by applying a combination of analysis of existing usage and user testing of fixed components, websites can be fixed in-flight and with great return on investment (ROI). As Agency's Munday says: 'Assessment of the site from a usability viewpoint doesn't have to cost a lot - but can return great value in the form of better customer retention and loyalty, conversion, and scalability.'

We mentioned ROI. It's one of the peculiarities of usability that it's actually quite difficult to separate out specific ROI directly attributable to improved usability on its own. 'It can be hard to quantify,' confirms SciVisum's Jones. 'Talk about ROI on usability is difficult, because few companies isolate it out,' admits The Usability Company's Blunden. 'But users tend to report to us that as part of a general revamp of the website they may see massive increases in turnover for as little as £40,000 worth of work.'

If you think 'usability' and 'user testing' is a weird combination, you probably won't be that worried about 'wasted', 'investment', and 'your website'.

Case Study: Victor Chandler Bets On Usability

Major bookmaker Victor Chandler (www.victorchandler.co.uk) says it's reaped tangible benefits from its usability improvement work.

Joe Coughlin, Gibraltar-based project manager at the firm, used The Usability Company to consolidate three different websites and improve overall usability - a process that took nine months, but resulted in what he says is 'at least' a 30% increase in business since.

Coughlin says a key reason to bring in a usability third-party is that in-house expertise is very useful, but you may still have blind spots as to what real-life customers actually do. 'While we know an awful lot about betting products, we discovered many ways we could make account management and registration go an awful lot smoother,' he told Internet World. 'We also found that online punters didn't always want to do things the way we wanted them to.' As a result, Coughlin is a convert to prototyping and user testing for all e-commerce sites. 'You'll save yourself an awful lot of time,' he says - and advice from your bookie is normally worth following.

Return to newsletter