Josh Ledgard and the team at KickoffLabs are doing some really exciting things for businesses big and small who want to quickly build and deploy landing pages that drive high conversion and registration. I particularly liked their recent emailed set of basic A/B testing recommendations that apply not just to landing pages but most destinations across your Web site.
I will add based on experience that testing testing testing (and testing some more) is the only way to consistently increase conversion. What’s more, the most arcane and counter-intuitive things might be what gives you the biggest leap in conversion. I’ll never forget testing dozens of variables independently on a landing page years ago, only to find that the single-biggest leap in conversion from a single variable was by changing the form background color from white to light blue.
Here, with permission, are additional recommendations from KickoffLabs.
Test Headlines
Your page may have a lot of copy. Don’t get stuck in the weeds A/B testing minute details. See if radical headline changes speak better to your audience than your baseline.
Test Big Visuals
Do you use big screenshots, a background, or a video? Try swapping them out. Test a video against screenshots to see what works better for YOUR audience. Remember that standard generalizations and best practices may not work for your customers.
Delete Stuff
You may not need everything on your page. Try removing some copy, images, extra buttons, links to Facebook, etc. If the variation with less stuff works just as well you can spend your time refining the details that matter.
Make it Pink
Does your call to action stand out enough? What if it was hot pink? You may be surprised by what you learn. 
Consider the Source
Try tuning your tests to the context of your visitors. Take a look at your referrals and figure out if they are coming from your ad, a tweet, a blog post, etc. Then try and test some messages for those people.
