If you want to know what’s working, test it. Is this recipe better than that one? Is this pair of shoes more comfortable than those? Marketing is no different. By taking a portion of your list and testing one element at a time, you can find out what works best. Even if you’ve done testing in the past, things change. It’s essential to keep testing to make sure you know what is working now.
Unbounce, a service that allows marketers to build, publish, and test landing pages, has described the impact of testing on three different companies and the lessons learned from each. Let’s take a brief look at each one.
Test #1: Great ads can get better. SafeSoft Solutions had a slick, professional ad that was doing well. The headline promoted productivity and efficiency. It used attention-grabbing, easy-to-read bullet points to outline the benefits of its services. But the ad did not contain pricing information. SafeSoft decided to test the addition of a green starburst with its pricing inside. The result? A 100% increase in conversions.
Test #2: Not all trial offers are created equal. For some companies, a seven-day offer might be perfect. For others, customers might need a more extended test-out period. When HubSpot split tested its trial offer, it found that prospects required more time to make a decision. By testing a 30-day trial over a seven-day trial, it was found that it could boost conversions by 110%.
Test #3: Location matters. Does the placement of the CTA make a difference? Inbound Strategy wanted to find out. It tweaked its site’s landing page, added more information, and played with the location of the CTA. The results? When the CTA was moved from the right-hand side of the page to the left, there was a 217% increase in conversions compared to the control.
These examples have lessons for us, as well. Whether you are working with print, email, landing pages, or any other channel, you're missing opportunities if you’re not doing split testing regularly. What insights and higher conversion rates might you be missing?
