How To Increase Landing Page Conversion Rates: 3 Rules For Effective Split-Testing
By Anna Johnson on August 2nd, 2010If you want to increase conversions on your landing pages you’ll definitely want to split-test those pages. By seeing how changes to key page elements impact conversions, you can make ongoing improvements and continuously increase conversions.
As simple as this sounds, though, there are a few ‘rules’ to keep in mind when putting together your tests. Here are three (3) rules for split-testing landing pages:
Rule #1: Clarify goals and metrics
The first question to ask yourself is: what are you trying to improve? Is it the opt-in rate, the ‘add to cart’ rate, the sales conversion rate, some other metric, or a number of metrics? Or is it something more general such as lifetime customer value or overall sales or profits?
Ideally, you want to work back from your overall goal (e.g. sales, profits, etc) to the page in question. Identify what main action you want people to take on that page, and how such action leads to accomplishing the overall goal. Then identify the metric that best represents the achievement of that main action.
If your landing page offers just one action for people to take – e.g. sign up to a list – then you’ll know what metric to measure i.e. the optin rate. If your landing page provides more than one action for people to take, then you might have primary and secondary goals and metrics to measure.
In any case, be clear about your goals and what metrics represent the accomplishment of those goals.
Rule #2: Segment traffic
Ideally, you should have dedicated landing pages for each traffic source. For example, an email campaign should drive traffic to a specific landing page, and a pay-per-click ad to another landing page.
This, however, is not always practical. In fact, you won’t always need separate landing pages for the purposes of identifying how different traffic sources respond to a given landing page.
You do need to be able to clearly identify and segment those traffic sources, though. This will allow you to see which changes to your landing pages increase conversion rates among different traffic segments. If there are significant differences, that may be reason enough to create distinct landing pages for each segment.
Rule #3: Take into account ‘upstream’ changes to your traffic funnel
Not only is it important to consider how changes to landing pages affect different traffic segments, but it’s also important to consider the impact of ‘upstream’ changes. Upstream changes are changes to elements of the campaign that brings traffic to your landing page in the first place.
Logically, a key change to the means by which you generate traffic (e.g. the type of email or ad you run) will affect the quality and quantity of traffic that reaches your landing page and, consequently, the conversion rate. For this reason, you want to prevent changes to upstream factors from unduly interfering with, or affecting, your landing page test results.
Alternatively, if you want to test upstream changes as well as landing page changes, just take care to when constructing both tests to ensure the integrity of both.
