All posts tagged website design

A/B Testing. If you have a website, read this article.

A/B Testing your website home page for effectiveness is essential. Call Dzined now on 01992 350066
A/B Testing your website home page for effectiveness is essential. Call Dzined now on 01992 350066

A/B Testing your website home page for effectiveness is essential. Call Dzined now on 01992 350066

If you’ve ever had an eye test you’ll understand about A/B testing.
Your optician will change lenses around until they find the most suitable type for your eyes. The optician will then continue with the “Is this one better or worse than the last?” approach until the optimum version is found. Job done!

Well there is an equivalent version of this available for your website.

Why should I bother?

The point of optimisation is simply to find out the most appealing version of your website from your site visitors perspective. Which version do they do what you expected them to do most often? You see you might think it’s obvious that they should just click here or use the big button your web designer added a year ago, but they might a have different opinion. So what’s the best version?

Well the “best” version may not be what you think it is.

If you go through a process of optimisation of your website home page (that’s always a good place to start but you don’t have to start there) you are looking for improvements in performance between 2 versions (version A and version B). Those improvements will be different for every business, so for example if you run an e-commerce website  it might be an increase in sales (or sales value perhaps) or it might be an increased sign up rate to your blog or more e-book downloads etc.

No matter what the goal of your optimization tests you are looking for the best ways of letting your website visitors reach the conclusion you want to reach without putting obstacles in their way to prevent them getting there.

So hows that done then?

There are several tools you can use but the one we would recommend you start with is the Google website optimizer tool. It’s very powerful, its easy to use, gives great information and best of all – It’s free to use.

Here’s a summary of how it works. You will need to have 2 versions of the test page designed with different layouts or different elements on the page. These “elements” could be the use of a picture vs no picture, or they could be colour vs black & white pages or perhaps a full on layout with tons of links vs a very simple layout.

The point is there are 2 versions being tested against each other and although we cannot predict which will perform best-  I guarantee you that one WILL out-perform the other. Just like the optician testing one lens over your eye against another, you will find that your website wasn’t optimised before and there IS a better way of doing it.

This sort of information is utterly invaluable because it could return a business to profit and make you think recession? what recession! If your business is down 10% on last year and you improve your conversion rate online by 10% then that could tip the scales in your favour for a change.

This is a very worthwhile exercise for any website owner because its extremely cost effective & quick to set up and get results. Google will split your site traffic into 2 and automatically send half one way & half the other. It’s also clever enough to set a cookie to make sure that the visitor always gets to see the same version of the page when they visit again so the results are more reliable.

The programme will look to send 100 visitors to 1 version & 100 to the other at least so depending on the number of site visitors you get each day, you can quickly calculate the time it will take to get some data.

How will I know when I have optimised my site?

Once you get the first result (lets say version B was best for the purpose of this example) you can set up another A/B test. You would keep the version B and make a new version A based upon B with more variations. As you get more information you might want to vary the elements within each page.

So you might have found that by adding buttons gets a better interaction rate and people do what you need them to in the first tests, you can try varying button colours or using images instead. If you found adding a title to a page worked best, then you can vary the title.

So initially you look for broad stroke effects, then you refine these to get the true optimal layout. Eventually you will see smaller & smaller differences in the results, so you can then start to look at other data like how people navigate through the site and start to optimise more pages & even the checkout process. Reduction in shopping cart abandonment can be like a pay rise on its own.

In the past we had to rely on our instincts or good old experience to run a marketing campaign, thats still valid now but (as they say) the numbers never lie and however clever you think you are (or your boss think he/she is!!) you now have an unprecedented opportunity to prove beyond all doubt what works & what doesn’t.

Sound great, where can I get me one of those?

Google website optimizer tool is available for anyone to use but if you would like more information (because there is LOTS more to discuss) on use of the tool or maybe some help setting it up or just some ideas on designs and layouts to test, just call dzined on 01992 350066 or email sales@dzined.co.uk

 

Home | About Us | Returns | Contact Us | spam free