71 Things to A/B Test

Share

Often times the hardest part of A/B testing is determining what to test in the first place. After having worked with thousands of customers who do A/B testing every day, one of the most common questions we still hear is, “Where do I begin?”

Conveniently, website testing inherently generates more questions than it answers. Your first test can lead to a whole litany of follow-up tests and iterations. For when you’ve exhausted all of those, or if you’re just getting started, here are 71 ideas (some more serious than others…) for testing your website.

Calls to Action

Your website exists for visitors to take action: reading, purchasing, signing up, downloading, or sharing. Here are some ways to test calls to action that can yield quick, easy, and big wins.

  1. Buy Now? Purchase? Checkout? Add to Cart? Change the call-to-action (CTA) text on your buttons to see which word or phrase converts more visitors.
  2. Try varying the location of your CTA button, making some CTAs more prominent than others.
  3. Test multiple CTAs per page against one CTA per page.
  4. Try using hyperlinks instead of buttons to find out which display your users prefer.
  5. Find out if CTAs with text, icons, or text + icons convert more users on your site.
  6. Try stripping the content off of your entire page. Replace it with a “Click Here for Free Beer!” button.

    Screen Shot 2013-04-30 at 1.29.02 AM

    Photo courtesy of The Optimizely Beer Club.

  7. Test different CTA hover states to make it more obvious that buttons are clickable and create a feel of interactivity on the page.
  8. Test different colors, shapes, and sizes for CTA buttons on your website.

Content

Content fuels your online business, particularly if you’re a B2B company. Testing how you position and advertise content on your site can uncover big conversion and engagement lifts:

  1. Test gated content against non-gated content. Find out if your users are willing to sign-up or provide more information to access materials on your site.
  2. Do site visitors crave more information about your company before signing up or making a purchase? Test adding or removing “About” content on your homepage.Screen Shot 2013-04-30 at 1.48.29 AM
  3. Content tone can make a big difference in keeping users on your site. See what your visitors prefer by testing various tones and styles.
  4. Test how your content is displayed. Do users prefer to scroll down the page or click through to another page to learn more?

Copy

Copy is your direct line of communication with website visitors – it sets the tone and helps users understand what you’re all about. Use these tests to make the copy on your site better resonate with your audience.

  1. TL;DR. Find out if your site visitors prefer shorter versions of headlines, taglines, product descriptions, and other content on your site.
  2. Test different headline text. Try variations that are straightforward against ones that are abstract, goofy, or creative.
  3. Test paragraphs
    • versus
    • bulleted
    • lists.
  4. Test how you frame your copy. Users may have different reactions to positive versus negative messaging.Screen Shot 2013-04-23 at 5.52.03 PM
  5. Try using haikus,
    For the product descriptions,
    Throughout your website.

Visual Media

Digital media has the power to greatly influence conversions and engagement on a website, and using AB testing tools to test it is a great idea because the right media can subconsciously influence people to act in a way that’s aligned with your testing goals.

  1. Test different types images on your landing page. People versus product is a good place to start.

    Screen Shot 2013-04-23 at 5.43.52 PM

    Source: 37signals

  2. Iterate from there. If your users prefer an image of people, test gender, age, number of people in the image, etc.

    Screen Shot 2013-04-23 at 5.44.15 PM

    Source: 37signals

  3. And iterate some more! How about trying a static image vs. a product video vs. a 360* product image.
  4. See how a stock image stacks up up against an image of your employees or customers in action.
  5. Test auto-play versus click-to-play videos.
  6. See how your product videos perform against this video of a sloth hugging a cat.
  7. Test a rotating carousel on your homepage versus a static image or video
  8. Try making your site easier to read. For example, test larger type, higher contrast colors, and non-comic sans fonts.
  9. Try putting a bird on it.
  10. Test different voice overs for the videos on your site. Test whether a male or female voice leads to the most completed views.
  11. Try different variations of your site’s product demo: animated versus screencast.
  12. Test a live-stream puppy cam on your homepage. All puppies. All the time.tumblr_me113zUjvO1r67nlm

Funnels

If your goal is to get more people from one page to the next – like in a checkout funnel, signup flow, or lead nurture – then A/B testing is your best bet. Funnels are rife with low-hanging fruit to test:

  1. Test removing extraneous distractions – like product offers, promotions, or shipping information – from each page in the purchase flow. Oftentimes a simplified experience can drive more conversions.Screen Shot 2013-04-30 at 3.07.43 AM
  2. Test the number of pages in your funnel. How does packing more information on one page compare to spreading information across multiple pages?

Testing Tip: If you are testing multiple pages in your funnel, make sure the new experience on your pages are unified by using a multi-page test.

  1. Test removing navigation to any pages outside the checkout funnel.
  2. Or try replacing certain steps within your funnel with modal boxes. For example, try making shipping options a modal box instead of a page. Screen Shot 2013-04-30 at 2.06.13 AM

Site Navigation

From the moment a visitor lands on your site, the navigation menu sets a foundation – it’s how people maneuver your site’s flow and prioritize what’s important. Here are some ideas for how to make it better:

  1. Test the order of menu items in your site navigation.
  2. Test the display of your navigation bar. Do site visitors prefer a horizontal or vertical orientation?
  3. Or what about a fixed navigation bar that travels down the page as your site visitors scroll?
  4. Test out the title of your navigation items. A simple change, like Why Use Us to How it Works may have a significant impact.

Testing Tip: If a test fails, try targeting the test to new versus returning visitors. Returning visitors are accustomed to seeing the site in a certain way—if a link or menu item is missing from the spot they normally go to find it, they’re not going to do the work to locate it.

Forms

Any potential friction point on a website is prime for testing. Forms are frequently cumbersome areas of websites. Try these tests on the forms on your site:

  1. Test the length of sign-up forms. Try removing non-essential sign-up boxes or relocating them to a page further down the funnel.
  2. Everybody loves free stuff. Try a special offer, discount, or promotion to increase sign-ups.
  3. Nobody loves spam. Try adding text that assures users you won’t fill up their inboxes with unnecessary junk.
  4. Try making individual form fields larger. Larger fields feel more friendly.
  5. Try asking for different information in your form fields. For example, business email versus email, or work phone versus cellphone.

Mobile Site

The mobile web is pervasive. Improving your mobile website through testing will help create an optimized experience that generates more clickthroughs, revenue, and conversions.

  1. Test the length of your mobile web pages. Are mobile users more willing to click to a new page or scroll down a page when browsing your site on their device?
  2. Try different displays and navigations options. Blinds, buttons, and blocks are a good place to start. Screen Shot 2013-04-30 at 12.33.44 AM

Testing Tip: When testing your mobile website, try targeting mobile users based on their operating system – Android or iOS, for example – to learn more about your mobile website visitors.

A/B testing ideas, ask questions and share success stories

Advertising

Split testing increases the value of the money you’re already spending on marketing programs, such as search engine marketing. To ensure you’re getting the biggest bang for your buck out of each paid initiative, try these tests:

  1. Test the headlines on your paid campaigns to see which ones get the most clicks.
  2. Try changing up the display URL on your ads. This can impact how many visitors click the ad.
  3. The landing page each ad directs to is an excellent place for testing. You paid for that visitor to land there, so why not do everything you can to convert them?

Still with us? popcorn Good.

Social

The reasons someone would share your site are many – make it easy for them to do so. Here are a few tests you can perform on a split testing tool to increase likes, retweets, and +1s on your content:

  1. Change the size and placement of social icons to see what compels users to share more often.
  2. Test standard social media icons against ones you’ve designed to match the look and feel of your site. Screen Shot 2013-04-29 at 11.18.10 PM
  3. Try finding your optimal Twitter voice. Tweet the same link with different of text at the same time two days in a row and see which tone of voice gets more engagement.
  4. Test different types of customer reviews on your site to see which are most compelling to your audience. Some examples include testimonials, Yelp reviews, and ResellerRatings.

Email

How do you make sure your marketing emails get opened and dare we say, clicked? Here are some testable elements that can increase open rates and clickthroughs:

  1. Test length and copy of your email subject lines.
  2. “Hey so-and-so!” Test personalized versus un-personalized emails.
  3. Find the optimal time to reach your audience by testing different days and times to send.
  4. Worried about scaring customers away with a newsletter? Optimize it! See how a weekly send stacks up against a monthly send. scare2
  5. Would readers prefer an email from your CEO, your marketing director, your broader team, or someone else? Test different “from” addresses to find out.
  6. Try varying the design of your email blasts, optimized for different desktop and mobile browsers.

Personalize It

Today, we’re more accustomed to web experiences that are custom-tailored to who we are and the URLs we’ve come from. Try testing these personalization techniques and see if visitors convert better.

  1. Create seasonal or holiday-specific promotional offers and test them on visitors living in specific locations.
  2. Test auto-filling form fields related to a site visitor’s location. Make things easier for your users.
  3. Test matching language-specific content to users coming from a certain country or region.
  4. Test different page designs and messaging for new versus returning visitors.
  5. Test whether showing different landing pages for visitors coming from mobile devices versus desktop browsers performs better.

Pricing and Shipping

Finding the right pricing point can drive more visitors to make a purchase. Use these tests to maximize revenue from your site:

  1. Test offering a free trial versus a money back guarantee to see which converts more users in the short and long-term.
  2. Test having checkboxes auto-selected as default. For example, a customer’s billing address could be defaulted to the same as their shipping address.
  3. On your pricing page, test whether annual billing or monthly billing generates more subscriptions.
  4. Try anchoring customers high before showing them a lower price. For example, “Competitors charge $2.9 trillion, but you can use us for just $2.99!”

Real World

Optimize your life.

  1. Fire up your Cuisinart and test your favorite toaster pastry: Pop-Tarts versus Toaster Strudel. It’s on.
  2. Host an old school Pepsi challenge
  3. Test your might. 
  4. Determine whether you’re left-brained or right-brained.
  5. Planning for a big office move like we are? Test new office chairs to optimize employee happiness.

    Greg and Julio enjoying a good, old-fashioned office chair A/B test.

    Greg and Julio enjoying a good, old-fashioned office chair A/B test.

These recommendations are a great starting point but are by no means a comprehensive or conclusive guide to optimizing your website. Did we miss anything? Let us know in the comments below!

A/B testing ideas, ask questions and share success stories

Share
comments powered by Disqus