But this is 2017. Surely, we have better tools than ever to unearth what it is customers want. We’ve never been better equipped to test web pages before they are rolled out. So why do brands continue to make a hash of launching a new site?

One basic reason might be the temptation to go for a big bang launch, complete with PR fanfare. Great if it works. But what if conversion rates suddenly drop through the floor?

You won’t have enough usable analytics data to identify where the problems are so you’ll either have to make changes and hope for the best, or quickly restore the old site. When you can make a series of controlled and tested incremental improvements, why take the risk of the big bang relaunch? That’s the riskiest thing you could do!

It’s interesting to compare the approaches of Google+ and LinkedIn when they relaunched. LinkedIn seemed to do a great job of annoying the hell out of some of its most important users by plonking the new version on their desktops without much warning (I’m referring to LinkedIn’s previous relaunch here, not the one currently underway).

These people shared, very publicly, what they didn’t like about the new version. As the roll-out gradually reached other users there was an expectation that they wouldn’t like what they were about to see – even though for most of us it turned out to be okay.

Google+, on the other hand, went out of its way to keep users informed. Google ran the new and old versions side by side for several months and people could switch back and forth at will. By the time the new version was fully rolled out there had been changes based on the feedback and there was very little outcry.

The BBC website is also one that seems to be in a constant state of development. It offers new options for keeping up with news, sports results etc., that you can try out, but always with the option of going back to what’s familiar. When new features are fully rolled out, users have been involved and everything is thoroughly tested.

Surely this is a smarter way to approach website upgrades and relaunches. Compare this to CNN which, in a desire to ‘update and refresh’, launched a site that used more resources and made it harder for readers to find the news that interested them – users hated it. Or how about the legendary Digg.com relaunch that almost killed the business.

Learn from your current site before relaunching

A classic mistake is to assume there’s nothing to learn from your existing site. Okay, it’s going to get binned. But you have thousands of customers using it every day providing data on what they want, how they want to do things and what they find difficult. You need to make use of that data.

Yes, it does make sense to do usability studies even on a site you are replacing. That way you can focus on improving the parts people dislike, and keep hold of the things you know they like and use.

And while you’re at it, talk to your customer service teams. They’ll have some excellent insights to offer on where people find the current website troublesome, as well at where there’s room for improvements to be made.

What does your business need to achieve?

Every business has targets: the number of new customers, sales growth by product/service category, lifetime customer value, cost of acquisition. How often do these business goals feed directly (and I mean directly) into your website redesign?

It’s one thing to launch a new website because you need to increase sales by 20%. It’s quite another to identify exactly how the new site and the activities that feed traffic to it will achieve that goal. And it’s yet another thing to have the test data to show that the new site will deliver the conversions you need.

Businesses rarely approach website relaunches with this degree of confidence. That’s because they don’t join up the dots between what the business needs to achieve and what the website is designed to deliver. And they rarely put those assumptions to the test before they launch. Result: disappointing return on the investment.

With clear goals and certainty about the weak areas on your current site you can focus the development priorities more productively. Are your current below-target sales because people struggle to select the right products, or because too many shoppers abandon carts before completing a purchase? It certainly helps to know.

What user experience do you want to create?

You’ve collected data and insights on current issues. You’ve blended these with the business goals you need to achieve. The next step is to define a user experience that will satisfy customers and deliver your goals.

What, exactly, do people need to do on your site? How are you going to make this simple, enjoyable and rewarding?

Draft a succinct and crystal clear statement for each key page across the website that defines the main objective(s) for your new, improved customer experience. Refer back to this constantly as you design and build the new solution to ensure you’re still focusing on your primary objectives.

What does your brand stand for?

A website redesign is an excellent opportunity to revisit your fundamental brand values. What do you stand for? What is it that particularly appeals to your customers?

What needs do you meet, what value do you create, and why do you do it better than your competition? What emotional drivers decide how visitors will act? Do they want to picture themselves as being more healthy, successful, in control, influential or contented? Or are they looking for something else?

This analysis will guide colours, imagery, typography, content and vocabulary. Your insights will help you create more powerful CTAs and better performing landing pages.

Here’s a great example of some content guidelines we recently came across from the team at uSwitch:

uSwitch tone chart

uSwitch tone chart guide: https://ustyle.guide/language/tone.html

Making it real

So now you’re clear about what your target audience think of your current website; you understand how the new site needs to perform, and how it needs to support visitors on their journey to becoming customers. What now?

Wireframes let you test the structure and navigation against defined user journeys. How obvious will each step be? Are there too many steps? You can design the prompts and help users will need at each stage. You can make better informed decisions about content, headings and CTAs.

Design visuals start to build a realistic picture of the look and feel of the new site that you can test against the business objectives and brand values.

Everything you design can, and should, be tested before launch on a variety of devices. There are great tools out there for usability and A/B split testing that will take the risk out of your new web pages.

The testing never stops

Launch isn’t the time to put your feet up. It’s a time to dive into the data and see whether all the hard work is paying off. It’s a time to be plotting tests and optimisation efforts to keep the metrics improving and to squeeze even more value out of your investment.

The digital world moves quickly. Technologies emerge, and your customers will be trying to outdo your user experience. Plan how you are going to stay ahead in the long term.