- Competitive Analysis
- Search Engine Optimization
- Pay Per Click
- Website Design
- Tracking & Analytics
- Email Marketing
- Social Media Marketing
- Video Marketing
- Franchise Marketing
- Case Studies
- Case Studies
- Home services
- Home Design & Remodeling
- Commercial Services
At this point in the great game of online marketing, I hope you know your baseline visit-to-lead conversion rate. Once you have this number, you can begin implementing tests designed to improve it. We call this process conversion rate optimization or CRO.
Conversion rate optimization is especially important when you’ve already made an investment to increase website traffic. Once you have sufficient traffic, improving your site’s conversion rate—even if it’s only by a percentage point or two—will give you an exponentially greater return on investment. Basic testing isn’t hard to do either; in fact, it can actually be fun!
The Power of A/B Testing
There are many types of tests you can do to improve your site’s conversion rate, but I recommend you start with A/B testing (aka split testing). A/B testing allows you to compare the effectiveness of two versions of a web page, marketing email, etc. in order to discover which one has a better conversion rate.
A/B testing is extremely useful for quick tests involving simple changes. Will an orange “Submit” button result in more web form submissions than a blue version? Test it! Which email subject line results in a higher open rate? Test it!
Think of A/B split testing like a footrace between two runners. In each A/B test, two competitors are running next to one another on the same track. They start at the same time and have the same conditions with which to compete against one another.
However, like a close footrace, sometimes it’s difficult to determine which of the two different options truly “won” the race. This is where we can use math and statistics to help us understand which option is truly different from the other. Don’t start freaking out if you haven’t taken a stats class in years, and there’s no need to brush off a dusty math book in your office. Today, there are many tools available to help you determine the winner of any given race—no matter how close the results appear to be!
6 Steps to Setup a Conversion Rate Optimization Test on Your Website
1. Identify the business problem you are trying to solve. Every test starts with the statement of the business problem you’re trying to solve. Problem – I need more roofing leads. In this example, our goal is to increase visit to lead conversion rate, or the number of conversions (web leads + phone calls) that we get for the same amount of visits to the website. As an example, if you currently get 10 leads out of 1,000 visits, your goal with this would be to get more than 10 leads out of 1,000 visits at some point in the future (post test and change).
2. Identify the one key input that you want to test. With A/B split testing, we have to pick one key variable to test. If you pick several different variables, you get into what is known as “multivariate” testing—something that will require a pretty advanced knowledge of statistics to determine which of the options truly made the increase in the conversion rate. You will want to select one key input to your test. Some examples of inputs are: calls to action, banner text, offer images, benefit lists, contact form addition, or phone number addition.
3. Identify your hypotheses. Once you’ve identified which input you want to test, you will want to determine the details of your test. If we’re trying to improve the conversion rate of a particular website page, one of our test inputs will be the current web page. The other input will be our adjusted page—an exact copy of the current webpage with one difference. These two inputs—the original web page and the test page—are going to be known as the “control” and the “alternative.” An example of this would be testing a webpage without a contact form (current/control page) against a webpage with a contact form (alternative page). Just like in a footrace, this is where you are going to want to determine what the outcomes could be, either one of the competitors will win or they will cross the finish line in tie.
4. Run the test. Now that you’ve got your control and alternative identified, we can actually set up the test to run! Google Analytics offers a great tool called “Google Analytics Experiments” that allows you to set up, execute, and evaluate A/B tests. Once you’ve identified your control page and uploaded your alternative landing page, it will help in measuring the results of the test. Google Analytics is just one of our recommended tools to help you increase your website conversion rates. You’ll start by navigating to the Experiments link in the “Behavior” section where you will be greeted by a screen that looks like this:
Google Analytics Experiments will now help walk you through all of the steps that you need to take to efficiently run your test and ensure that you are testing in a controlled environment.
5. Evaluate the Results. Now that you’ve set the test up, you will want to allow the test to do its magic and begin collecting data from your visitors. You don’t want to check on your data every day, but you can check on it probably at least once per week. Google Analytics takes all of the guesswork out of trying to figure out which of your two competing web pages has “won.” Google Analytics will display what the status of the experiment is and all you have to do is go in and check to see if there is a winner.
6. Did you solve the business problem? Step six probably sounds like a no-brainer, but it’s amazing how many companies get through some portion of the steps above and never follow through. You must stay focused on the business problem at hand. We want to run the experiment until we’ve found a real winner and then implement the change that has won. The goal has been reached and you’ve solved the business problem!
In the event that you run your test for an extended period of time (>2000 visits), you may still get that there is no winner. When this happens, it means that there is not a statistically significant difference between the two options that you have presented and it’s best to declare the original web page, the control, to be the winner and begin another test.
Need Help Improving Your Conversion Rate?
Don’t waste your limited advertising budget sending visitors to a website that can’t close! Contact the conversion rate optimization experts at Blue Corona today for a free consultation and evaluation.
About The Author: Tyler is an all-star account manager in our NC office. In his free time, he enjoys training for ultra-endurance events, including marathons and Ironman triathlons.
View more blogs by Tyler Yost