Every few years, companies decide to redesign their sites. Sometimes they’re driven by a change in messaging, a new strategy, or product release. Sometimes they don’t like the old site anymore. Maybe the company’s leaders have changed, received negative feedback about the site, or seen new ways of doing things. In many cases, there are good reasons to redesign and update the site.
Often, there’s a small problem. We all know the famous saying: “I know half my advertising is wasted. I just don’t know which half.” Meaning, some of it works, some doesn’t. Still, we can often decipher what works and what doesn’t. Online, this is much less of a problem when tracked and measured successfully against your goals. But there’s still an area where this creeps in and causes problems: site redesigns.
Step back and think about a site redesign. You scrap what you have and start over in a number of areas. Some things may stay the same, but a lot changes. When you finally launch the site, you’ll have fixed a few things that weren’t working well. The chance of creating the best possible page, story, call to action, and so on are incredibly slim, but you’ve improved some things.
Often, there were many things working well on the old site that don’t work as well on the new site. Many times, they aren’t obvious things and won’t jump out at you from your behavioral Web analytics measurements. Rather, they’ll be apparent in attitudinal measurements (surveys, satisfaction, usability studies, etc.) combined with your behavioral data. They might not impact conversion, but they could affect offline behaviors and long-term perceptions about your company and brand. Remember: negative changes aren’t always easy to identify, but they can have a big impact on the customers and prospects who visit your site.
Recently, I was talking about site redesign issues with a senior member of our optimization team. He shared a few interesting analogies.
Redesigning a site, he said, is similar to buying a new set of expensive kitchen knives. The person researches and shops around, then drops $500 on a set of knives. After doing this, most people expect the knives won’t require sharpening the day after such a costly purchase. He also likened redesigning a site to buying a new car. You expect everything will work fine when the vehicle’s new and don’t think about problems you might have down the road.
Companies spend many months and many dollars redesigning sites so they’ll launch on time and on budget, thinking everything will be better and nothing broken on day one. But often, this isn’t the case.
How can you keep this from happening? There are a few key things that can help reduce the chance that your site redesign is actually a step back:
- Understand your current site. Make sure you truly understand what works and what doesn’t on your existing site. Measure things from both a behavioral standpoint and attitudinal. Create a performance baseline for your site’s key areas.
- Test throughout the redesign process. Test some ideas throughout the redesign process on the existing site. That doesn’t mean these changes will perform the same on the new site (when everything else changes), but it does provide a good indication of what will work and what won’t. During this phase, your greatest lessons will often come from something that bombs — and that you had thought would work well. Don’t test just a few things; test a lot of things in a lot of different areas. It will help increase your site’s success when it does launch.
- Develop a launch plan. Consider an A/B test when you launch the redesigned site. Roll out the site to a small portion of your audience. For the first week or so, only let 5 to 10 percent of your traffic see the new site. This will allow you to truly compare what’s working and what can be improved so you can address those issues right away. Have a team waiting to address the opportunity areas quickly, then roll out the site. You’ll hear horror stories about companies that launched new sites and their key conversions dropping on day one, but they can’t go back to the old site because it was scrapped. And they didn’t have a testing or roll-back plan, so they ended up just scrambling. Obviously, you must understand how this will impact your site visitors. There are ways to do this well, and there are ways to make this really awkward for you and your site visitors.
- Build testing into the site launch. There are always some obvious things you should test as soon as the site launches. Plan for them ahead of time and launch with those tests in place. On day one, you can hit the ground running with tests that you can review, then roll out the best performing. This also allows you to continually test those key areas during your site’s life.
- Create an ongoing testing strategy. For your new site, be sure to have a plan for testing that will carry on throughout the site’s life. Using behavioral targeting and testing tools, you can begin to customize the content based on what you know about your audience and their behaviors.
Don’t fall into the same traps that many companies do when redesigning their sites. Taking a little extra time to plan out your strategy can almost guarantee an improved outcome upon launch. Remember: listen to your customers, understand your audience, create the appropriate strategy, and tell the story experientially and visually in the best possible way if you want to be successful.
Emily Ma, product director of Tencent’s advertising platform products department, was a keynote speaker at ClickZ Live Shanghai where she discussed the ... read more
In today's multichannel world how can marketers use data to ensure the experience a customer receives is relevant to them?
The terms that customers type into your site search function can help you to gain an understanding of user behaviour and can be used to optimise ... read more