Those who’ve worked with me know that I’m not a fan of website redesign projects. I much prefer testing and incremental change.

Why? Because I’ve been involved in website redesign projects that were catastrophes, and many of my clients have admitted that they engaged my expertise in conversion rate optimization because they redesigned and their new site converted worse than their old one!

But, once in a while, a redesign is the right course of action. Maybe it’s part of a re-branding effort, maybe it’s simply a “political” requirement, or maybe the current site will require so much work to optimize that it’s not cost-efficient.

If you find yourself heading down the path of a redesign project, there are factors that your team can focus on to ensure that the redesigned site does more than just look nice in browsers on devices. It needs to perform as part of your marketing system and help your company achieve business goals! Building these 7 factors into your project will set you up for success.

I recently raised these 7 factors to a client who I believe needs a radical overhaul of their existing site design:

#1: Analytics, also known as “Being data-driven”

Web analytics data needs to be a major component in understanding what isn’t working about the current site, traffic patterns, and much more. Make sure you’re tracking, and have a baseline of data on, all the business goals of the site. Also make sure that your data is clean, meaning you’re filtering out internal traffic, known low-quality traffic, etc. A solid audit of your overall web analytics data will feed into most of the remaining 6 factors. This is often a good opportunity to bring in outside consulting help to pore over your data and present you with findings that will inform your data-driven redesign process.

#2: Design Considerations, a.k.a. “Who we’re redesigning for”

Before you redesign, you should have a good idea of who you’re redesigning for. For example, what is the breakdown of screen dimensions the design should work well with? How many pixels wide should the new site be? Where is “the fold” for your audience? What browsers are most important to optimize for? You can see now how web analytics data is crucial in this process.

Another major factor to consider is how much of your traffic is coming to the site on mobile devices and tablets. Depending on that ratio, and on trends for mobile traffic, you may want to make a fully-optimized mobile site, an app, a few landing pages, or do nothing at all…for the time being.

#3:  Visitor Intent, a.k.a. “What is the angle of approach?”

Beyond knowing the technical characteristics of your audience (#2), you need to understand the psychographics, or intent, of your target audience. What are they looking for when they find you? What problems are they looking to solve? How ready are they to make a purchase decision?

Again, web analytics data will be your starting point. Look at the ratio of new/return visitors and what channels drive the bulk of your traffic. Also, look at the keywords that drive traffic. Can you infer some customer intent behind those keywords? Of course you can! It’s valuable to know how “ready to buy” your audience is when they reach you, because that answer will drive factor #4…

#4: Content, a.k.a. “What are our prospects hungry for?”

Once you have an understanding of the intent of your visitors when they reach your site, you can determine what kinds of content the redesigned site should have, where content should be located, how it should be positioned, and more.

For example, if you determine the bulk of your visitors are in “comparison shopping” mode, which is often the case, how can your content be redesigned to fill their needs? Does your guarantee factor in heavily? Probably. Do videos and articles explaining the basics of your space need to be prominent and featured? Maybe not. Just try to map your prospects’ likely questions and needs to content that will satisfy them.

#5: Competitive Landscape, a.k.a. “Optimize your positioning”

Doing a redesign without doing competitive analysis is marketing suicide. Sorry to be blunt, but what good is a shiny new site if it isn’t properly differentiated from its competitor sites? Consumers of all kinds are bombarded with choice, so how will your new site stand out from the crowd?

Competitive analysis can take many forms, and can range from informal to highly structured. I would start by first figuring out who the competition is, i.e. Primary Competitors and Indirect Competitors. Study their sites in terms of messaging, visual language, content, and features. Find their strengths and weaknesses. Guess what? This feeds right into #6…

#6: Unique Value Proposition, a.k.a. “Why do business with us?”

I’ve written about Unique Value Proposition (UVP) in the past, and I continue to believe that it’s a crucial aspect of any online marketing optimization effort. For definition’s sake, I’ll defer to MarketingExperiments‘ wording of a succinct statement of the uniqueness of a business that sets it apart from all competitors.

If you believe you have a UVP for your business/site, a redesign project is a good opportunity to revisit it in light of your competitive analysis findings. If it’s still bullet-proof, think about how that UVP is going to be expressed throughout the experience on the new site.

If you don’t have a UVP developed, don’t bother redesigning until you’ve got one. Every site should have one, and over many tests I’ve found that adding or improving the expression of UVP has strong influence on winning test results.

#7: Optimization & Testing, a.k.a. “What and how to change”

Again, based on web analytics, you’re going to have a good idea of which pages on the site are pivotal; which ones are where conversions will/won’t happen. You’ll also know which pages have to do a lot of persuading in order to drive conversions. These pages should be launched with a “backlog” of test hypotheses so that you can start the continuous improvement process after 2-4 weeks of benchmarking.

Based on past traffic data, you can know which pages on the site will have enough traffic to test (e.g. major entry pages, homepage, etc.) and which ones should just be launched with your “best guess” and be revisited down the road (e.g. About Us page, site map page, etc.)

As much as possible, the new site should be architected with testing in mind. This may mean that certain pages live outside the Content Management System and/or aren’t database-driven. Generally, if you code to modern standards, most testing tools should “play nice” with the code on your new site.

Finally, I urge you to do some testing before your redesign to vet some of your ideas about things you want to change. For example, if you’re planning to redesign the global “header” of the site template, you can test different designs on your target audience before you invest in coding and deploying them. I’ve even seen this done with new logos! Another example is page layout. You might test whether a 2-column or a 3-column “grid” is most effective, and then base redesign decisions on test results.

Conclusion

I’m almost certain I’ve missed a factor or two. Any suggestions? Have any of you out there gone through this process in an optimization-centric mind set? Been through a catastrophic redesign process? Drop a comment!