If you use the web to market your business, then you may have heard the phrase A/B Testing. You might be wondering what is involved in A/B testing. In the world of online advertising, it means that an advertiser creates two landing webpages or two web advertisements that you test between to find the best performer. This takes the guesswork out of determining the best return on investment using data. Often I recommend gathering statistics between two pages to determine performance only “by the numbers”.
In most cases, you will only want to test between two pages or ideas. To find significant performance increases the pages should differ in an extreme way. You can run two pages side by side in subsequent tests with minor differences to understand slight increases. Let’s consider one example of a test.
I worked with an insurance company that was looking to get more leads. We put up a page with a simple form and a few bullet points of text. Then we created a second page that had a form, a few text points, a photo of a young lady who we named “Gwen” and we put a video on the page.
After we had purchased ads from Google and Facebook, we sent them to a URL that had an A/B testing tool code on it. The code switched between A and B pages for about 1,000 views. Our clicks from Google cost us about $3 each, so our testing budget was $3,000. In some cases, you can find traffic sources and keywords in Google that only cost $0.25 each. A budget helps you to not get emotionally attached to your test and instead only run your campaign for what is actually needed to get the results you are looking for.
At the end of the test, we found that the video was played more often, but the page without the video got more lead form fills. After 3 months, we looked at the individuals who filled out the form. The data showed that a majority of people who watched the video became insurance clients, however, the lead form gave more leads for the sales team to follow up on. We assumed because the video educated them that they were more prepared to sign up for insurance. If we had a need for a larger volume of leads for the sales team, then the non-video page would be the winner.
Some other types of pages and offers I have tested between include:
– Free Sample Page
– Free Consultation
– Online Party Invitation
– Video Tutorial Registration Form
– Paypal Product Purchase
– Request for Information Form
– Video / Infomercial form
– Webinar Registration
– Video Series Registration
– Headline A/B Testing
– Images Testing
– Button Color/Size Testing
When you are seeking to test between two pages you will need technology to serve the pages evenly. Let’s look at a few tools that can do the heavy lifting:
You can create a personalized message to website visitors by leveraging the optimizely platform. They show one website where the background changes based on the words that brought the visitor to the site. A Florida visitor will get a different experience than one who responded to a special product offering in Canada. This is a great way to segment and personalize to customers.
I recommend the entry level plan and scheduling a test of 500-1500 visitors. You will need to put a small piece of programming code in your html header on each page. Then you create a new experiment and a guide takes you through all the setup steps.
Optimizely Test Results Report
This A/B testing provider also offers personalized web experiences for visitors, geotargeting, and customer feedback. They do not setup using code but they are more expensive. There is a free trial available for you to check out what is really involved here.
The tool allows you to trigger between pages and to analyze what goal you will measure by. They also have a heatmap for each page which allows you to see what a majority of visitors click or mouse over. This company realizes that there is a learning curve when it comes to A/B testing and they have written a guide to help you understand the steps, A-Z, in order to create a successful test with their platform.
This A/B testing tool delivers templates that are tested and thematic. It also reports analytics in A/B testing for these lead driving web pages. The company says that their tool gets you higher conversions, creates pages fast, requires no installations, and requires no IT personnel. Unbounce is a cool tool because it allows you to change things on the page without coding. Here are some of the Unbounce features:
– Mobile Responsive Designs
– Landing Page Drag and Drop Builder
– Form Builder (No HTML Knowledge Required)
– 85+ Landing page templates
– Leads go to any email
– Integrates into WordPress
– Dynamic Text Replacement
– Script Management (Cut and Past Google Analytics codes and other codes onto the page)
– Secure Encryption
– Real Time Stats
Here is an overview video of Unbounce.
Any of these tools might be the right one for you. Once you use one of these tools you may decide to run 3,5, or 10 more tests each year to hone in on your conversion rates. One test could provide a lift in your conversion rate that will give you a higher return on your investment dollars that you spend on email advertising, pay-per-click, and social media ads each year. I hope that this article has inspired you to take your marketing to the next level. Check out some of our other tool overview articles for Analytics, Blogging, and more!