As a final post in our six part series on Conversion Rate Optimization, this entry will step back and take a top-level look at the steps involved, and sequence in which they should be performed.
1.) Setup and Customization of Tracking & Measurement
The first step in assessing how you can up-level your performance is to ensure that you have the proper tracking in place. This will give you an accurate picture of your current performance, as well as performance of tests you’ll run in the future. At minimum, you should track:
- Conversions — Track macro-conversions — sales, leads, sign-ups — whatever the end goal that you want your visitors to perform.
- Conversion Funnel — In addition to tracking conversions, you’ll want to track the sequence in which users arrive at that conversion event. For example, if users go through a checkout process in order to buy products on your site, you’ll want to set a custom conversion funnel that tracks behavior through those checkout pages. This will make it easier to identify pain points within this funnel.
- Behavior Metrics — These are actions that don’t directly correspond to conversions, but give an indication of the degree to which people find your site relevant and usable. For example, track: bounce rate, time on site, and average number of pages viewed per session.
2.) Gather Baseline Data
Before you start testing, you need baseline data against which to assess the performance of your tests. If you’re just now setting up your tracking, as mentioned in step one, wait for at least one month with full tracking before you assess and establish baseline data.
3.) Use Baseline Data To Establish What To Test
Once you have enough data to get a sense of how your website is performing in it’s current form, use that data to make a list of tests you’d like to perform. Base these tests on:
- Pain points you’ve identified in your conversion funnel
- Pain points you’ve identified on your landing pages
- Website components for which you’ve identified additional variations to test — calls-to-action, design and formatting, messaging experimentation
4.) Run Your Tests
After you’ve established which tests you want to run, it’s time to launch!
Keep the following in mind:
- Only run one test per website element at once — For example, it’s ok to test the removal of a sign-in requirement from the checkout page at the same time as experimenting with a different landing page design because you’ll likely be assessing results based on two different sets of metrics. However, removing the sign-in requirement at the same time as adding another field to the checkout funnel simultaneously will make it harder to understand which checkout funnel test made the difference in performance, since you’ll likely be assessing results by conversion rate — which both tests would directly affect.
- Establish a timeline for tests — …and stick to it. Run each test for a specific amount of time. Do not stop tests prematurely, based on the assumption that you have enough data to take action. This can result in an asymmetrical comparison between elements that you’ve tested separately.
- Don’t Alter Tests –– Don’t change your content in the middle of the test. This will void the results because you’ll be looking at data for two different things in the end. If you need to make a change, abort the test, alter your content, and begin a new test, subject to your testing schedule.
5.) Take Action Based On Your Results
After your tests are complete, you’ll be able to compare performance against your baseline data and make your implementations. Make sure of the following:
- Your findings are based on a statistically significant amount of information
- You’re drawing the right conclusions, based on your end-business goals
- You’re not finding patterns where none exist. Make sure your findings are actually backed up by the data.
6.) Rinse and Repeat
Always keep testing! This should be a constant process by which you’re making continual improvements to your website and online marketing efforts. Adopt Conversion Rate Optimization as a frame of mind, not a one-time opportunity to make some improvements.
Conversion Rate Optimization is the process of increasing conversion rate through testing. Follow these steps in sequence:
- Setup and Customization of Tracking & Measurement
- Gather Baseline Data
- Use Baseline Data To Establish What To Test
- Run Your Tests
- Take Action Based On Your Results
- Rinse and Repeat — Always keep testing!
References and Resources: