7 AB Testing Best Practices

What is one of the easiest ways to test website variations to determine which design will provide better conversions? The answer lies in these A/B testing best practices. In using A/B testing, half of a site’s visitors will see one design variation. The other 50% will see a different variation. The results are then compared to determine which design provides a better overall conversion rate.

This tool is invaluable when it comes to optimizing a landing page. When implemented properly, A/B testing will minimize the amount of time and money that is spent in creating the best landing page possible. Consider these best practices on your next project to avoid making mistakes that could make things worse instead of better.

1. Have a Theory To Test In The First Place.

Many web designers wind up creating dozens of different buttons, graphics, and other calls to action just to experiment with what will work and what won’t work. This is like trying to find a needle in a very large haystack. You might be able to stumble onto something fantastic. Anything is possible. Having a theory to follow where possible outcomes are expected often reduces the size of the haystack that requires searching.

The goal here is to have an understanding of what is being tested and why it needs to be tested. Change for the sake of change isn’t what is best for business. Inform everyone involved about the goals that each change is expected to create, work towards meeting that goal, and you’ll be able to have a more comprehensive look at the information that has been created.

2. Subtle Changes Track Better.

A/B testing is often used to compare two completely different landing pages to determine which one is better. That’s like asking people if they like apples better or if they like oranges better. Some people will like both. Some people won’t like both. You don’t really get any information beyond the fact that one page might outperform the other page.

Subtle changes are what need to be tested during this process. Instead of comparing apples to oranges, try comparing Gala apples to Honeycrisp apples to see which one is preferred. When you do this, you’ll even get information from those who like oranges more because you’re gauging specific interest in two subtle changes.

3. Don’t Give Up On The Testing Process.

Many landing pages stop testing once they receive a conversion rate that satisfies their goals. The only problem with this is that goal satisfaction and conversion rate maximization are two very different things. Having a 20% conversion rate is pretty good, but what if they conversion rate maximizes at 40%? You could be losing 100% of your conversions and never even realize it because you’re satisfied with the results.

Not only is it a best practice in A/B testing to test early, but you also should be testing often and through an ongoing process. Your landing page can become a little bit better every day. Even a one word subtle change in your call-to-action content can add percentage points to your conversion rates. Never be completely satisfied.

4. Give It Time.

The biggest mistake that is made in A/B testing is that landing page optimization happens too often. Meaningful results are not going to be discovered overnight. You’re tracking patterns of change instead of immediate results. Step back for a moment, let the changes mature over the course of about a week, and let your visitors show you what changes they like and which ones they do not like.

Take a deep breath. Now take another. Maybe take up some meditation. You’ve got to be patient when it comes to A/B testing because impatience could cost you conversions.

5. It’s Not About You.

The point of your landing page isn’t that it should look “awesome” to you. Your landing page isn’t about you at all. It’s about how it plays with the visitors to it who are interesting in what you’ve got to say. That’s why in the A/B testing best practices, it is always a good idea to keep an open mind when deciding what works and what doesn’t work.

It’s a common trap. We all get locked into certain routines when it comes to marketing and landing page creation because our experience has shown us that specific things work and specific things don’t work. The only problem is that the original ideas that we had before might not apply to this new targeted demographic. Becoming locked into certain design pathways is an easy way to chart a course toward failure.

You can’t predict the behavior of your visitors. You can, however, be in control of your behavior. That’s why adjusting to the sometimes unpredictable actions of your potential leads will keep your conversion rates a lot higher.

6. Allow Yourself To Be Imperfect.

Every A/B test is going to be different. This means the approach to each test should be unique. The results that were achieved before are not going to be achieved again. Sometimes imperfect landing pages perform better than the ones we all think are refined and perfect. A strategy that works for one business might not work at all for another business.

The point here is that if you are too specific with how you enter A/B testing and examine results, then that inflexibility could wind up costing you conversions. You don’t need to have a perfect test to discover results that are invaluable to your efforts. Even if just one small change is being evaluated, there are always going to be hundreds or thousands of influences that shape the results from external sources.

Here’s the truth: the “perfect” landing page is a myth. There is always something that can be improved.

7. Keep Up The Good Work.

If you have run out of ideas for your landing page and your optimization results are better than you could have ever believed, then it is time to choose another landing page to begin testing on your site. A website is a lot like a house. There are many windows and doors that all allow entry into the building. Some people will come through your bedroom window. Others will walk through the front door. Your website’s bedroom window is a landing page for that visitor and it needs to be just as optimized as your front door option.

When you move onto A/B testing for your other pages, it is important to remember the lessons you’ve learned over the first optimization process. It is also important to forget them. Each page is unique and must be treated as its own entity. Your demographics could react very differently to your bedroom window page. You must be ready for that possibility in order to be successful.

These A/B testing best practices can help you achieve better conversion results over time. Keep testing, never be satisfied, and don’t be fooled into thinking perfection has been achieved. You can always be a little bit better today than you were yesterday.