The problem with web design, is that everyone uses websites and as such everyone has an opinion. This can sometimes be more of a hindrance than a help when redesigning your site as everyone’s opinion is valid but whose opinion is right?

This does raise an interesting question – who should have the final say on the design of your website?

• The client – Head of marketing?
• The client – Product manager?
• The client – Sales manager?
• The client – HiPPO – ‘highest paid persons opinion’?
• The agency – Web designer?
• The agency – Project manager?
• The agency – Web developer?

Scroll down to see the answer…

>
>
>
>
>
>
>
>
>
>
>
>
>
>
>
>
>
>
>
>
>

Answer: None of them.

The person whose opinion is most important is your customers. After all they’re the one who’ll actually be using the site, right? And who you’re trying to persuade to part with their hard earned cash?

So surely it should be your customer’s opinion you should really listen to the most?

Multivariable testing
Easier said, than done? Multivariable testing allows you to get this input and feedback from your customers without the need to contact them or ask them to fill-out an annoying survey or questionnaire (who answers them honestly anyway?).

Multivariable testing works by running several versions of your design side-by-side, live on your website* and then measuring which version performs ‘best’. The crucial point though is that ‘best’ is something REALLY tangible like the ‘number of sales or enquiries’. Not something subjective like – I think most people prefer blue, it’s a much nicer colour.

The end result is cold hard facts – blue buttons give us 46% more sales than green buttons, FACT. Who in their right mind would go with green buttons?

How does multivariable testing differ from usability testing?
Usability testing allows you to test your site design on real people i.e. your typical users. But this is typically in a lab like environment and with just a handful of humans (between 2 and 10 typically) and as such the opinion of just one person could have a huge impact on the results.

Multivariate testing on the other hand lets you test your design with 1000’s of your customers, so you get a far more accurate picture from loads of people rather than just a handful.

We’re not recommending that you should skip usability testing if you’re opting for multivariable testing – the two complement each other really well and are aimed at different stages of the design process.
What we are saying is that crucial decisions about the design and layout of your website can be tested in field, on your live site – without your customers even realising it.

Sounds amazing, how can I learn more?
Check out some of our other blogs on the subject or visit our new website dedicated to website conversions.

* For the purpose of this blog I’ve oversimplified the ‘how it works’ bit, but only to make the concept easier to understand as an introduction.