Category: UX advice / Research articles

Work out how your current website performs before designing a new one (aka why study history)

The Forum in Rome

Here’s a question I’ve received from clients that is worth a decent answer:

“If we’re aiming to completely redesign our website, why are we studying what’s happening on the existing one? We don’t like it and want to change it altogether!”

The wrong way round

It seems like a fair question, but ultimately it’s built on the wrong way of thinking about a website and its audience. Users tend to behave based on their needs, rather than the way the website is designed.

Assuming your business isn’t changing altogether, your new website is going to be targeting the same set of users as your existing one. If you’re offering the same product people will come to the site with the same expectations, and will be looking for answers to the same questions.

It’s useful to study the past to learn how people have acted in similar situations and to help solve today’s problems. As the oft-quoted saying goes, “those who fail to learn from the past, are doomed to repeat it”.

A complete redesign?

You may think that you want to ‘completely redesign’ everything or re-imagine the way it all works, perhaps coming up with ‘fresh’ and ‘innovative’ designs. I hear this from clients a lot, but it’s not truly what they want. The answer is rarely to throw out everything and start from scratch.

What most business-owners actually want is to get rid of the failing version they have and increase conversion. Many (naturally) jump ahead of themselves and assume a complete redesign is the answer.

If we re-orientate ourselves around the goal of increasing conversion then our approach looks a little different. When we already have a website offering the same product to the same set of users, we can find out what they’re aiming to do and why we can’t meet their needs at the moment.

How to

We can study users’ existing behaviour through analytics, session recordings, heatmaps, and of course user testing. Through this we can see what they are really looking for and what they‘re ignoring.

It might turn out that a complete redesign is what we needed. Or it might just be a new page or two fixing the biggest problems. If we can back up our process with evidence and can show how it will improve conversion then the business should be happy.

(For more on exactly how you can do this yourself, take a look at my framework for redesigning any website. It’s explained further in my online video course, The Evidence-Based Redesign.)

Don’t ignore the best asset you have for seeing what your users need and in your enthusiasm to make something new, put your design before their needs.

Of course if you’re considering changing your strategy or the business model altogether then you can take a different approach. When the fundamentals change then you really might need a complete redesign—it can be worth doing a process like a design sprint, to quickly discover the right solution.

Last updated on 6 November 2019
UX Design Templates Multipack

Articles on similar topics

An evidence-based framework for redesigning any website

How to write remote user testing tasks (that don’t bias your results)

Why most designers can’t do UX