The RIGHT way to split test your online campaigns.

The RIGHT way to split test your online campaigns.

Conversion Rate Optimisation (CRO)

How do you know for sure if a change will help or hurt your business? What if you and your partner have differing opinions on a marketing campaign? How do you maximize your Conversion Rate (conversion rate optimisation)?


Conversion Rate OptimisationHow to use Conversion Rate Optimisation


The only way to know for sure is to test. Yes, even tried-and-true classic marketing rules can be found to hurt certain businesses once they are tested.

How do you test a theory and know for sure that was the single factor that created the improvement or hurt sales? Split testing. You run ads to two or more different versions of whatever you are testing at the same time splitting hits to all versions equally. This is the only way to tell accurately how the single change affected the behavior of your audience.

For example, if you ran one version for 2 weeks and the another version for 2 weeks after that, you won't know how the timing affected the campaign. Were there more competitors running ads during one of those campaigns? Was there a shift in seasonal demand during one week vs. another?

Therefore, you need to control all other variables so you can see if that actual singular change made a difference in conversion. You could test money back guarantee effect on sales, different versions of a capture page, or different versions and length of forms.

For example, you buy Facebook ads to the same audience to 2 different versions of a capture page where viewers opt-in for a free ebook. Once each page has had enough viewers that the results become statistically significant with a 95% confidence interval (software like Optimizely can tell you when that occurs), you can see which page has the most conversions and shift 100% of your campaign to the higher converting opt-in page.

It’s a good idea to run campaigns and test for a 2 week period at least to hit every day of the week twice. It can also be a good idea to get 100 conversions on each version before making a final decision. If there are multiple choices and some of the choices are clearly under performing, you can disable those choices early and run traffic to the remaining contenders only.


Here are some tools to help you run these tests:
Leadpages Landings Pages
You can create split tests (also known as A/B tests) using an existing Leadpage or Template.

ClickFunnels Landing Pages
Optimize your funnel conversions through A/B split testing
“Test and personalize across your customers' journey with Optimizely, the Experimentation Platform.”
Create and A/B test different versions of your website to continuously discover the best performing versions that increase your online sales.
Form Analytics shows you how to increase conversions
Find your hottest opportunities for growth today with screen recordings, conversion funnels, form analytics, feedback polls, surveys, user testing, heatmaps, and more.
Landing pages

Share on:Share on Facebook0Share on Google+3Tweet about this on TwitterShare on LinkedIn2Share on Reddit0

Leave a Reply

Your email address will not be published.