Conversion rate optimization can be a huge addition to any company’s digital marketing efforts when done correctly. From increasing click through rates and decreasing bounce rates to increasing on site goal completions, conversion rate optimization (or CRO) can play an integral role for your business when it comes to creating a site that performs well.

With so many case studies around the web discussing how companies leveraged CRO to increase overall conversion rates to make more money than ever before, it can be tempting to jump in and test everything on a website without thinking about the goal of such tests. Not everything on a website that can be tested should be tested. Doing that can actually cause more issues for a website than it does to increase a company’s bottom line.

In an effort to make sure that you are testing the right elements of your website, here are 4 tips to correctly use CRO for your business in order to get the most ROI our of your tests:

1. Look at bounce rates

Out of all the metrics to take note of when deciding whether or not a CRO test on an individual page makes sense, bounce rate and exit rate are the most important. Take a look at the pages that have the highest exit rates and compare them with pages that have a low exit rate. It makes sense to leave pages with low exit and bounce rates alone as they are performing as they should be. However, pages with high bounce or exit rates are often the source of lost traffic (note that individual blog posts are clearly exempt from this general rule) and are sensible pages to test if they have large amounts of traffic coming to them.

When analyzing a page with a high bounce or exit rate, ask yourself why this is happening. Often there is content that is missing from the page that audiences were looking for and the page just doesn’t provide it, or they can’t find key navigation elements to take them where they want to go. This means taking a look at key CTAs to make sure they are clear to audiences both in terms of where they are on the page and what kind of copy is featured.

2. Test the Obvious

Headlines and CTAs are the best pieces of a website that you can test as they will ultimately have the biggest effect on user behavior on an individual page. Changing the messaging, color, or size of your headlines or CTA’s are also great ideas for A/B Testing as they are the main focus point for users who are trying to get to the ever-important RAQ or product pages.

By testing elements that your users are going to be using to get through to your key goal completion on a site, you will get to test the most helpful elements that affect your bottom line. Tweak these elements to the point of being blatantly obvious so that users don’t have to guess what they are supposed to do next.

3. Use User Surveys to Inform Your Tests

A great way to make sure that your CRO tests are going to have the biggest impact possible is to understand what users need or are looking for before you even start CRO. Using user surveys or user testing tools is a great way to understand why your website is not performing to its full potential. When using user surveys, make sure to use qualitative questions in order to get the most out of it! Don’t make your questions make the process seem like a net-promoter score survey for users. You want to avoid leading users into simple “yes” or “no” responses, and instead aim to ask open ended questions so that you get specific feedback about what you can do to make your website work better for your audience.

Dive into the responses you get and explore any potential trends in responses that users are giving about your website. If you notice patterns in terms of the feedback suggestions you get from your audience, it makes sense to test them in your CRO tests. This way, you’re sure to be testing things that your audience actually wants or needs instead of basic elements that users might not even be noticing in the current version of a page or website.

4. Don’t Micro Test

Testing little elements on your website one by one is often not beneficial enough for your website and testing timeline (especially if you have low traffic) to be worthwhile. Some websites, like Google, can get away with testing micro-elements because they have millions of people using their site, so it takes them less than a day to get statistical significance on their tests. However, for sites that only get a couple of hundred visitors a day, it could take months to reach statistical significance for tests that don’t provide that much ROI for a business. That’s why it is so important to test only the elements that are going to have a big impact.

The founder of Optimizely and author of A/B Testing, Dan Siroker, states that it’s better to make big drastic changes and test it, instead of small changes. Once you have found a big change that works, you can go back and analyze the next steps of how to optimize that change and understand what worked in order to get the most out of the new element that you’ve found.

Don’t Lose Sight of Your Goals

It can be tempting to wildly test a variety of different parts of your website and justify doing it by telling yourself that regardless of what happens you will have some interesting data that you can repurpose into case studies about what you’re doing or what you found. However, it is important to not lose sight of the whole point of a CRO study, which is to help increase your bottom line by testing elements that you or your users are having trouble using properly. Keep it in perspective and follow the tips I’ve outlined and you’re sure to find new, interesting ways to increase your website’s ROI.

Do you have tips I’ve missed or questions about CRO that you would like addressed? Let me know in the comments section below.