The Master Marketer BLOG

Website Design Post-Launch Testing

Posted by Jeffrey Schmidt on Tue, Nov 06, 2012 @ 6 PM

Welcome back! We're continuing our series on website redesigns today to talk about what is possibly the single most vital period whenever you're creating a new website design: the post-launch testing period.

Just because your new website has launched doesn't mean it's time to relax. Here's how to make sure you get the most out of your redesign and ensure it really is living up to all the hopes and dreams you had a few months before.

Pay Close Attention to Criticisms

Any mail sent to your complaint or tech support in-boxes should get special priority and significance. If more than a couple people complain about an aspect of the site being hard to use or -worse- not working right at all, you need to get on it immediately.

Also, keep involving your customers. After the new website design launches, ask for more feedback on your blog forums and social media accounts. Look for areas you can tweak to improve the user experience based on feedback.

Website Split A/B Testing

Landing Page Testing: A is A. Unless it's B.

You'll want to pay close attention to the performance of your Calls to Action (CTAs) and landing pages in the wake of a full site redesign. The great thing about Landing Pages is you can tweak them whenever you want to try to get more conversions.

The best way to go about this is through A/B testing. All this means is that you have two versions of a landing page, and you let them both compete against each other live for a week or two to see which one performs better. Plenty of high-end CMS systems like Hubspot include this feature, and Google has a free Content Experiments system that allows for the same thing.

  • Proceed scientifically: Don't make changes willy-nilly. Always have a control, and always change a minimum of elements at a time - preferably only one - so you'll never be in doubt as to what changed your conversion numbers.
  • Give each change at least a week: Fundamentally, the longer you run any one experiment, the more reliable the data will be. The only time an experiment should be cut short is if it gives truly disastrous results.
  • Keep backups of everything: You never know when you'll hit a blind alley and want to "back up" a few steps to try something different. Keep your old experiments around in case you spot something in one you want to try again.

Don't Drown in Data

We occasionally refer to the "data maze" you can get trapped in when you start getting deep into SEO analytics and metrics. Don't get so lost in your data that you lose sight of the proverbial forest for the trees.

Fundamentally, there are only a few metrics which are really important:

  • Page visits: Everyone loves page visit numbers. Just remember, while this is a "sexy" stat to tout, it doesn't make you any money directly. Don't panic if your numbers following a redesign peak and then fall off for awhile; that's normal for a hype cycle.
  • Lead conversions: You want your CTAs and landing pages operating at maximum efficiency to keep your sales database full.
  • Sales: 'Nuff said.
  • Bounce rate: This is the dark cloud to go with your page visits' silver lining. If a high percentage of your visitors leave again in less than a minute, you need to revise your content or presentation.

And that just about does it for our website design series. Tune in next time for a wrap up, as well as some advice for monitoring the performance of your new website in the future.

Call DeepSky Marketing to find out how you can create profitable marketing strategies for your company. DeepSky Marketing is a company that provides businesses with profitable marketing systems and verifiable return on investment (ROI). To schedule a brief no-cost consultation call 707 823-3888.

Posted by Jeffrey Schmidt

Subscribe to our weekly newsletter

Get it in your inbox ;)

By entering your email you expressly consent to receive our newsletter every week and other material related to DeepSky Marketing.