In my two previous posts I have explained why I believe the web design industry (as we currently know it) is in for a major shake-up during the next few years.

No longer will it be good enough to build websites that look nice (essentially works of art). Web designers in the future will need to think like marketers, first and foremost. They will have to think less about being “creative” and more about selling.

Yes, selling. That’s what makes the world go round in business. One of the oldest adages in business is: “Nothing happens until someone sells something.” Yet this appears to have been forgotten by the “creative” experts in the web industry.

Nice-looking sites… hardly any sales!

Even the few web designers who do think like marketers (myself included) have tended to rely on making an educated guess about how to design a website that will sell effectively. We’ve used the experience gained from previous clients, in the expectation that the same principles will work for a different client in a different market.

Sometimes, we strike it lucky. Our client’s new website starts selling like gangbusters. They tell us how wonderful we are… and our ego grows bigger. These sites get added to our portfolio and we write case studies about how our websites have dramatically increased our clients’ businesses.

But equally often, the results of even the most experienced, marketing-driven web designer will be disappointing. The site will be slow to build up traffic and will not bring in as much new business and the client hoped for. These are the sites we want to quietly forget.

Sometimes, a plain web page outperforms a glitzy one

Often, extra effort put into prettying up a website’s graphics has no effect at all. I am continually amazed when I meet people who are quietly making a fortune from a crappy-looking website. The reason is, they’re smart marketers. They know how to make an offer their website visitors can’t refuse.

Could these people do even better with a prettier-looking website? Maybe. But more often than not I’ve seen conversion rates actually fall away when the website has been redesigned.

Creativity + analysis + success

I’ve become more and more convinced that web design should be a process that occurs over a period of months, rather than simply building a site, launching it and then leaving it to sink or swim.

We need a process of testing and tweaking, so improve the website’s conversion rate.

We need to begin with…

1. A clear goal in mind (a certain number of sales, enquiries or some other action we want our website visitors to take)
2. A way to measure that goal.

In other words, we need measurable results. Increasingly, clients will demand this of web design companies. Much as it strikes fear into our hearts, it’s a truth every web professional needs to face, sooner rather than later.

A strategy for optimising websites

I’ve recently begun testing some of my clients’ websites using Google Website Optimizer.

Basically, it works by designing two slightly different versions of the same web page – version A and version B. You might change the headline, or the placement of certain elements of the page, or other changes.

Then you “split test” version A against version B using Google Website Optimizer. This means you send 50% of the traffic to version A and 50% to version B. The whole process is trackable and measurable, so you can see which version of the web page produces better results.

But you don’t have to stop there. You can then come up with versions C, D, E etc and test them against the “winner” from the previous test.  As a result of this process, you come up with a web page that is the most effective.

It took me a long time to accept the need for testing

I first heard about the principle of split testing in 2004. Back then, a few of the top American Internet marketers were doing it. But I resisited the idea I could actually do any split testing for my clients.

Firstly, it would involve extra work. My clients would need to pay for my extra time, and they would complain, I reasoned. Furthermore, my clients expected me to know what would work, first time around, without having to test it. It would make me look incompetent if I admitted I didn’t always get it right first time.

So I shelved the idea of testing. I continued to build websites that pleased my clients. But the niggling doubts increased until I recently (late 2009) decided to take the plunge and embrace testing.

So, do we still need “creative” people in the web industry?

Yes, of course! A good creative process gives you strong alternatives to evaluate. If a web designer lacks creative skills, or is lazy, they will simply end up testing “bad version A” against “bad version B”, which will only give a bad result.

The way to build world-class websites is to start with the best you can find, and then to test them against other great alternatives.

No longer will it be good enough to design competitive websites simply as a “creative” – because creative ideas and graphic design are only a fraction of what a website needs to be great.

In the coming new world of web design, we will still use all our creativity to develop several great solutions. But that will be just the start. Then we’ll put those solutions to test… and find the winner!