Why Multivariate Testing is Essential to Optimizing Conversion Rate

Conversion Rate Optimization: Why You Should Test

– Testing your website content and design is essential for conversion rate optimization. Webpages have multiple elements, each of which can have limitless designs or different positions on the page. Each element, its design, and position can play a role in how well a particular webpage can work in the conversion process. Finding the right combination of elements is why you test.

Testing Methods

Experienced marketers use a couple of methods for conversion rate optimization testing: A/B or multivariate.

conversion rateA/B testing is pretty simple. You have two versions of the same page. The first page might have a red background and the other a blue background, but everything else is the same. You funnel half your flow into one page and half to the other. You measure how long people stay on the page and how many interact with it (i.e., click the “Buy Now” button). The one that gets the best results wins. You go through the process multiple times. This can help with conversion rate optimization, but not as much as multivariate testing.

Multivariate testing takes A/B testing to the next level. Instead of testing a single element on a page, you test multiple variants on the same page. You set up multiple webpages with all the different combinations of those factors and funnel visitors to each one in roughly the same numbers and test their preferences. The results will help your conversion rate optimization efforts.

A Simple Example

Let’s say you want to test out the background color and the position of the “Buy Now” button. You set up four pages to test for conversion rate optimization:

  • Page #1 has a blue background and the button on the middle of the page. Twenty-two (22) percent of visitors to this page hit the button.
  • Page #2 has a red background and the button on the middle of the page. Twenty-three (23) percent hit the button.
  • Page #3 has a blue background and the button is at the bottom of the page. Thirty-nine (39) percent hit the button.
  • Page #4 has a red background and the button is at the bottom of the page. Forty-one (41) percent hit the button.

Getting Insights

conversion rateThe biggest advantage of multivariate testing is the insights you get. These insights are valuable for conversion rate optimization.

Using this example, the most preferred page was #4 with the red background and the button at the bottom of the page. So, your audience prefers red to blue, and buttons at the bottom to those in the middle, right? Actually, the only real preference they have shown is on the placement of the button.

The visitors were essentially indifferent to color. Comparing the pages with the button in the middle (#1 and #2), there is only a single percentage difference. Comparing the pages with the button on the bottom (#2 and #4) there is only a two point difference, which will not really help in your conversion rate optimization efforts.

The real difference comes when you compare the placement of the buttons. Comparing the blue pages (#1 and #3), you see a jump from 22 percent to 39 percent when the button moved from the middle to the bottom of the page. Comparing the red pages (#2 and #4) shows a similar jump when the button was on the bottom instead of in the middle. That kind of jump is critical to conversion rate optimization.

Multivariate testing offers powerful insights into your audience’s preferences. You can learn more about conversion rate optimization?and ways to enhance your website in this ebook:










Marketing Matters Inbound?is an inbound marketing firm specializing in online conversions, content?strategy, PPC campaigns and traditional media placement. Our experience in?B2B digital?marketing?and?traditional media?allows us to deliver results through a strategic media mix. We believe in?aligning and merging marketing and sales/operations for optimal return on your marketing investment.

0 replies

Leave a Reply

Want to join the discussion?
Feel free to contribute!

Leave a Reply