Landing Page Testing: How to test and what to test

By ,

106 SHARES

Testing your ad copy and your landing pages can significantly improve your paid search efforts. Of course, building a solid keyword base, creating an optimized account structure, and executing a well-planned bid management strategy are also crucial. However, testing allows you to understand how to optimize and improve your communication with your target audience. Conducting thorough tests on your landing pages can deepen your audience interaction and increase your conversion rate.

If you’re like a lot of companies, running tests on your landing pages used to involve numerous meetings with your IT department, development hours in order to get everything designed and launched, as well as additional time to analyze the active tests. Your life as a search engine marketer got easier when Google launched Website Optimizer. If you have access to your landing page code, and you have some basic coding skills (or someone in close proximity does), then you can quickly launch tests, analyze the results and adjust accordingly.

With Website Optimizer you can run all of your testing in one location. (And no, this isn’t a paid plug for for this tool. We just think it’s extremely helpful!) You can monitor your tests, make adjustments as needed and improve your conversion rate. As far specific technical requirements are concerned to launch tests with Website Optimizer, I’ll leave that to the help section of the official website.

However, once you have your account open and your ready to start improving your landing page performance, what exactly should you be testing? Here is a list of landing page/website elements that you can test in order to learn what appeals best to your audience.

Headline: Your headline, along with almost every other element on our landing page, needs to be relevant, timely and appropriate for your audience. First, your headline has to assure the user that they’ve landed (pun intended) in the right place. In this vein, your headline needs address the core concern of someone who arrives on your page. From the first second of a user’s arrival, you need to tell them that you have the answer to their search.

There are numerous ways to write and test great headlines. Here are just a few ideas for testing new headlines:

  • Try using emotional copy that will appeal to the user’s hopes, dreams, fears, aspirations.
  • Test customer-focused vs. company focused headline (example: We can make your lead prospecting easier vs. You can make your lead prospecting easier).
  • Split test using questions against declarative sentences.
  • Try using longer headlines that are loaded with benefits against shorter headlines that focus solely on setting up the conversion.
  • If you have headline copy that works well, then you can test the different font sizes and color.

Body copy: This is where you back-up the claims made in your headline or PPC ad. If your PPC ad inspired the user to click and the headline has convinced them to read further down the page, then your copy needs to do the heavy lifting.

Your body copy also needs to be relevant to a user’s search, benefit-driven, and written in the manner in which your audience expects. By this last point I mean that you need to know how colloquial, formal or technical the language on your landing page should be.

For the purpose of improving conversion rate, here a few body copy elements to test:

  • Try using short text against longer text.
  • Test copy that is more emotionally driven, rather than technical or feature driven.
  • Your copy should always be relevant, but you can test inserting more of your high-traffic keywords into your copy.
  • See if a list of benefits helps increase your performance.

Call-to-action: By this I could mean a few different things: the phrase that you use on your landing page for the desired action, or perhaps the button that users click on to go to the next step in your conversion funnel. Either way, you need to test it.

In regards to actual text, you should test a few different phrases to see which appeals to users best. Does a shorter call-to-action such as, “Sign Up’ work better than a more detailed one such as, “Get Your Free Guide Now.”

Also, you can test the color of your call-to-action buttons. Usually when there is an article written about landing page testing, someone mentions the button color. But it works and it’s worth testing.

Contact form: In the past, I have found that the next two are the hardest to test. The difficulty lies in the fact that the contact form is usually tied to a database. And if the contact form is damaged in anyway, then the leads will not populate correctly. However, if you can adapt, then this can be a quick win for testing.

For contact form testing, you should try out:

  • Different lengths of your form. Try to using a longer form, and then try a version that is short enough to get above the fold.
  • Ask fewer questions. How much information does your sales team need to follow up with a lead?
  • How many required fields do you have? Can you make some of them optional?
  • In regards to contact information; do you display your phone number on the landing page? How many calls do you receive from your landing page? Try removing the phone number to see if this inspires people to fill out the form.

Trust/Credibility symbols: Does your industry have certain certifications that will display your level of expertise and help build trust with your audience? If you don’t have them on your landing page, you should test this out. Also, if you’re accepting any type of payment, displaying safe-purchasing symbols can help improve your conversion rate.

Try out different offers: This is a quick one: you can highlight seasonal  or time-sensitive offers on your landing page. This is pretty straight forward. See which offer generates the best response and use it again at the same time the following year or even the next month.

Mini-site vs. Landing page: Once you have conducted a series of tests on your landing page and you feel that it’s as good as it’s going to get and you’ve hit the point of diminishing returns, then it might be time to go back to the drawing board. By this I mean you may need to completely re-think how you use landing pages. However, at least when you start this process, you’ll have a control landing page that you can test against.

With this tactic, you may to test out a mini-site or a multiple-step conversion process. Or if you’re using one of these longer forms, you may want to go the opposite direction and use a shorter landing page.

If you want more ideas on how to optimize landing pages for higher conversion rates, you can check out our podcast series, PPC Hero Landing Page Optimization Podcast.

For more in-depth information on landing page testing, I highly recommend Tim Ash’s book, Landing Page Optimization. And I also recommend Always Be Testing by Bryan Eisenberg and John Quarto-vonTivadar for additional information on Website Optimizer. Very helpful resources!

Keep in mind that there is a cost to landing page testing. Some of your tests are not going to be successful, but your results should improve over time. Optimizing your landing page is a continuous process that will lead to enhanced results when executed properly and with care.

Get more weekly links with our Fast Five newsletter! Five Fast Links in Your Email Every Friday.

Also send me a daily RSS digest

Automation Toolkit

Twitter Facebook LinkedIn Google+ Email Print More
  • http://rbeale.com Ryan Beale

    Hi Joe,

    Nice Post! One thing that I would add to this post is, NO NAV MENU on the Landing Page. If you have not tested a landing page with NO NAV MENU, please do so and let me know the result. My bet is that your conversion rates will go up. :)

    RB

  • http://www.vizad.com Robert Brady

    Great information Joe. I was just wondering what your thoughts were on landing page software like LiveBall. Have you had a chance to use it?

  • http://webstrategyworkshop.com Andy Brudtkuhl

    I’d have to agree with Ryan on the no navigation menu. You don’t want to give potential conversions an easy way to leave your landing page. You sent them there for conversion – not to look around on your site :)

    Other than that – great tips. And ALWAYS BE TESTING :)

  • http://www.pcchero.com Joe

    @Ryan and @Andy: Thanks guy! It’s funny… most of the landing pages I work with already don’t have the navigation included so I wasn’t even thinking about that! Good point!

    @Robert: I haven’t used LiveBall but I’ve heard good things. If you try it out, let me know your thoughts!

  • http://theppcblog.com Matthew Umbro

    Hi Joe,

    I work at a company that specializes in user interface design so I tend to have our Information Architects create PPC landing pages. I have been very pleased with the pages they create for me and in turn have seen very good conversion rates.

    I’m happy that you are writing about landing pages because it is often overlooked. Setting up a structurally sound campaign is necessary, but only half the equation. If you are not sending your users to a relevant landing page you are throwing away your money.

    Other items that are often overlooked, but can improve your conversion rates include:

    Appropriate Tabbing Structure
    - making sure a user can hit the tab button to get to the next field)

    Privacy Policy
    - a line of text at the bottom of a form that assures users their information will be kept confidential

    Browser Capability and Screen Resolution
    - making sure your landing pages are accessible in most browsers and screen resolutions

    Keep up the good work!

    -Matthew

  • http://HackCircle.com Web & IT security

    Thanks a lot for the insightful advice about landing page and improving landing page conversion rates. I am not sure if I should be using copy that appeals to emotions. I would rather stick with non-chalant business copy for the time being.

  • http://www.alanmitchell.com.au Alan

    Some great tips Joe. You’ve covered pretty much every angle of landind page optimisation. If I could choose only one element to test, though, it would have to be call to action buttons. Finding the right colour, message and placement of your call to action button(s) can make a huge difference for little effort.

    I think elements like credibility symbols would arguably have a smaller effect on user behaviour, so a large amount of data would need to be collected before it’s possible to tell one apart from the next.

    Guess the smaller and more discreet the change, the more data is needed for a significant result.

    Alan

  • http://www.ioninteractive.com Megan

    This is a great overview guys!

  • http://Www.Twitter.com/conversion_guru Rob Jackson

    I’m very suprised to see that no one has mentioned the use of analytics data to build tests. A solid methodology of analysis to drive a/b and mvt tests will result in experiments that are much more likely to succeed. Gone are the days when the tactic is to throw a load of ideas at a wall and see what works. Check out Google’s conversion room blog to see what I mean.

  • http://www.performable.com Rex Dixon

    Great stuff as always! Did you happen to sign up for Performable or share any tests over on A/B Tests?

  • http://www.templateminisites.com Shah

    I wish getting over a broken heart can be so easy as following a few steps.. but its not? :(

  • Levi Page | Stat Centric

     I’ve been using Website Optimizer and find it a lot easier than GWO.

  • Pingback: Are You Buying What I’m Selling: Landing Page Copy | Doppelpager

  • Rajesh Rao

    Thanks a lot

  • Pingback: Free Download Daily Update