Landing Page Testing

The Best Practices In Testing Landing Pages

The key to ensuring that your marketing strategy is effective is by tracing the performance by testing landing pages. Through using Google analytics, or other third-party analysis packages, we have already briefly touched upon four useful metrics that one should monitor on their landing page.

While the introduction article addressed what metrics to test, this article will cover what to consider in testing landing pages. How should you conduct metric testing on your page?

Test one variable at a time

It is important to reiterate that  testing landing pages gives out aspects that you can change on your page to improve performance. Generally, most of the testing will take the form of AB testing . Two versions of the landing page are compared to see which one performs better.

The key to any form of comparative analysis is to understand the difference between the test samples . It is unwise to have  two versions having multiple different elements as you will not know which differences caused the changes in the performance.

In a scenario where the two versions have different headlines, color schemes, image selection and call to action placement. It will be very difficult to know which changed elements to attribute any improvement on landing page performance.

The Best Practices In Testing Landing Pages

You need to test one aspect at a time. The image above shows the A/B testing for two different call to action designs. If version B achieves better conversion rates, you can confidently attribute it to the new vibrantly colored call to action button.

Afterwards you can change the other elements, one at a time, as you fine tune the most effective landing page design.

Have a clear strategy on the metrics being tested.

There should be a clear strategy when it comes to testing landing pages. Each landing page, depending on its purpose and design, has different elements. Therefore, you need to know beforehand, which variables in the landing page that you will be testing.

One of the best ways to know which variables to be tested is to have a clear performance gap. The exact one you are trying to address with the landing page testing.

What is your landing page not achieving? You might get enough visitors, which cannot convert  into leads. Such a scenario focus your testing efforts on the lead generation form design and/or the length of your landing page.

Sometimes you might have a very high bounce rate, then  focus on de-cluttering your landing page. Alternatively try different visual aspects of the landing page to see which design appeals most to your visitors.

The guideline of Landing page Testing is a clear strategy to ensure that it is effective.

You should have a control version

One of the biggest mistakes marketers make in testing landing pages is not having both versions being tested online simultaneously. When testing two landing page designs, both of them need to be online (and active) at the same time.

It creates a control version that makes it much easier to have a comparative analysis of the metrics seen from both samples. When you test different versions of the landing page at different times, other time related variables (such as differing inflation rates) may affect the metrics observed.

Have clear metric performance thresholds

The final step in testing landing pages is to have a clear cut metric performance thresholds. If you have identified that your landing page has a conversion rate problem, what is the ideal conversion rate that you are targeting?

If you want to achieve a 20% conversion rate through re-designing your lead generation form, then if you make changes that give you a 17% conversion rate, you need to go back to the drawing board to try to achieve your target.

Your goals need to be realistic and attainable. If they are, then you need to ensure that you try as much as possible to meet them when doing testing landing pages.

Without these clear cut goals, then your testing does not have a measurable performance outcome, which makes the effectiveness of the testing process difficult to ascertain.