“The website just feels tired.” “It doesn’t look new and trendy.” “We need to look more like our competitors.”
These are some of the phrases companies use to justify a redesign of their website. They always come with an assumption that new = better.
An old rule of thumb is to overhaul a website every four or five years. In fact, you can use the Wayback Machine to look at older versions of popular websites. At one point or another, those businesses sunk a lot of money into a one-time overhaul.
Design is an important part of raising conversions. KISSmetrics has 100 case studies that show how redesigns have boosted conversions significantly. But that doesn’t mean you have to discard what you’ve already built just to optimize conversions.
The Old Method
Radical (or Revolutionary) Site Redesign is the process of redesigning a website in one big overhaul. One day the site is just as you know it, the next it’s completely different and you’re hunting for your favorite content. It creates confusion for your customers and users.
Too much change in a short period of time makes it hard to track the influence of those changes on your conversions because you’ve adjusted too many variables. For example, your new layout might be keeping users on the page longer, but the opt-in form asks too many questions. You couldn’t know which is affecting conversions if you change both at the same time.
If your website depends on SEO authority to stay relevant, radical redesigns can tank your rankings, especially if you disrupt URLs, remove pages, and negatively affect your technicalities like responsiveness and site speed.
The worst problem with radically redesigning your website is that it’s hard to undo. Even if people stick with you through a big change, they’ll abandon your site if you decide to jump back to the original form. Overnight, your reliability will be ruined. Plus management won’t be pleased about spending tremendous resources on a redesign just to throw it away (even though this is the sunk cost fallacy, people still do it).
On top of all that, there’s always the possibility that the new design doesn’t lead to more conversions. Your “improvements” could have the opposite effect: less conversions; less revenue.
Some of the most popular websites on the web have injured their traffic (and ultimately their brands) by deploying sudden massive redesigns. Digg’s lost 26% of their web traffic in 2010. In 2013, Yahoo Mail redesigned their interface twice, creating mass-confusion. Their own employees wouldn’t use it! A CNN overhaul was met with a lot of resistance, as well. Fans reported fewer headlines above the fold, clumsy navigation, and (the worst sin of all), slow loading times – up to 21 seconds!
If any of these companies had A/B tested incremental changes, they would have gained valuable information about how their visitors used the websites without driving away traffic.
In many ways, Revolutionary Redesign is an arrogant way to do business. It doesn’t build a website with much feedback from the users, relying on aesthetics and basic heuristics. The end result is a web property that looks beautiful, but has a questionable conversion rate.
Evolutionary Site Redesign
Evolutionary Site Redesign was coined by Chris Goward, founder of WiderFunnel. It is the process of using smart A/B testing strategies to create incremental updates in your website that lead to increased conversions. This redesign process is best used when a brand isn’t broken (meaning you don’t need a whole new overhaul for PR reasons), but you want to make improvements.
ESR is often overlooked by business owners because it’s not sexy. There’s no big reveal of a new product. There’s nothing to share on social media or write a press release about. Releasing revolutionary changes does net a small bump in traffic, but it doesn’t last forever. ESR changes improve conversions forever.
Compare these two images, courtesy of WideFunnel. Even if a revolutionary website redesign made a website successful the day it was deployed to production (that’s a BIG assumption, but let’s just assume), the intervals between redesign periods are full of under-achievement.
ESR, however, is conversion optimization on a large scale. It transforms your website into something entirely new, but the new form is based on data and genuine user feedback, not the gut feeling or preference of some c-suite or designer. It keeps your website improving along a steady trend so you make money while you tweak the site.
The Benefits of Evolutionary Site Redesign
1. ESR focus on analytical data and ROI
At the end of the day, conversions are the goal, regardless of how the website looks. Would you rather have a beautiful website that doesn’t make any money, or an ugly website with high conversions and strong revenue? ESR constantly uses analytical data (not our assumptions) based on how people are actually using the website to push the conversion rate.
2. ESR leaves the pieces that work alone
Not every component of your website is broken. An overhaul of the entire site would likely disrupt the things that already work well. ESR keeps these pieces together, which is important so you don’t alienate your current visitors.
3. There’s little argument over whether something works
With ESR, decisions are made by testing different options against one another and comparing data. If one page’s layout converts 7% more traffic into buyers, no one can argue that it doesn’t work. Jim’s favorite type of layout or Tina’s favorite button color become irrelevant.
4. Success becomes more likely over time
Incremental changes actually become easier as you make more. Each time you make a change, you learn something. Eventually you’ll have piles of real data to consider before you design tests. For instance, there’s no reason to test purple buttons if purple has always caused a reduced conversion rate in the past. When your insights are better, your tests are more likely to lead to positive conversions.
ESR strives to reach an optimal performance of your website without undergoing big changes. But how do you know when your websites are performing at their maximum potential (what Andrew Chen calls the “local maximum”)? How do you know your conversions are as high as they can be?
You can identify maximum conversion rate once you begin to experience diminishing returns. That is, once your tests are only reducing the conversion rate. When you notice this, 52 Weeks of UX recommends stopping your optimization tests and “return to other kinds of analysis to figure out the next steps.” You might conduct customer interviews, perform surveys, perform a technical analysis, or do user testing.
If your website or application already performs in some meaningful way (that is, it creates some conversions and/or generates some revenue), a radical redesign isn’t just ineffective – it’s dangerous.
Improving on a continuous basis will keep your website modern and effective. It will even let you explore modern design trends as long as you base your decisions on data.
Effective Experiments is the perfect platform for large-scale conversion optimization. If you’re ready to begin Evolutionary Redesign, request a demo today.