CRO, or conversion rate optimization, is a complex process, and it’s easy for businesses to get overwhelmed.
With the number of metrics to compare, reports to analyze, and testing to do, it can feel like you’re getting overloaded with information. This can lead to CRO mistakes that cost you valuable website visitors.
The primary goal of CRO is to find out the truth about how visitors respond to your website and use it to improve business. Basing these business decisions on inaccurate information from mistakes made throughout the CRO process is unhelpful, if not detrimental.
Here are the top five most common CRO mistakes that beginners make (and how to prevent them).
1. Your test is too short
If you research A/B testing or ask any business professional, they’ll tell you to run your test until it reaches “statistical significance.” However, this can be a confusing term for those that are unfamiliar.
Statistical significance means that you are confident in your decision to choose one variation of your A/B test over another based on the evidence. If your test is too short, you have no choice but to base your decision on incomplete information.
To determine how long your test needs to be, you need to look at how big your sample size is.
This is basic statistics: The bigger your sample size, the more accurate your average values will be. The bigger your sample size, the longer your test needs to be to accommodate that sample size.
When calculating how long the test should run, you should also consider how many views per day the test page(s) receive. If you have a page that receives 50 views per day, and you need 150k views to achieve your 95% confidence level, it could take you over eight years to determine whether the test was successful.
2. You’re not testing pages that are doing well
The phrase “if it ain’t broke, don’t fix it” doesn’t apply in CRO. Everything can be improved, even if it’s already doing well. Marketing assets that are already working have the potential to perform even better if you take the time to optimize.
For example, say your website has old blog posts that have been influential in generating traffic and leads. Influential blog posts build up search authority and rank highly on search engines, generating traffic for years after you posted them. But that doesn’t mean your work is done; you will still want to convert people once they find your popular blog post.
The optimization goals of these blog posts should be to:
- Get more leads from these blog posts
- Increase traffic to other blog posts on the site
You can accomplish these goals by rewriting CTA’s (calls-to-action) or adjusting elements on each page to make the site look more appealing. A/B test two versions and see which your customers prefer.
3. You give up after “failed” tests
There is no failing in CRO unless you give up or something breaks. If you don’t get the results you want in your test, at least you learned something. Giving up when you don’t get the results you want means that you aren’t applying the knowledge you learned from the “failure” to improve your process.
Approach each test with the same mindset: Results are results, regardless of if they’re the ones you wanted.
Take the knowledge you gained from the “failed” test and analyze the numbers. Revisit your test to examine the reasons why it could have failed, such as a bad hypothesis, traffic channels, or visitor types. Once you’ve determined the reason for the failure, revamp the test and try again.
If analytics don’t provide proof to justify testing again, try to analyze the original page using tools such as heat maps and site polls. These will help you know where visitors are clicking and the frequency. Polls allow your visitors to give honest feedback about what they see. In turn, you’ll walk away with information about what visitors expect from your site.
Taking time to revisit your failed tests allows you to understand why the test failed and keep that knowledge in mind for your next test.
4. You make changes not based on results
Changes on your site should come from first-hand research. Most teams are keen to make CRO changes based on industry-known practices instead of testing their site and learning what strategies work for them.
Making changes not based on results from research is risky; you’re applying ideas that could work, but there’s no way to prove it. As a result, you’ll waste time you could be using to A/B test and collect valuable data. This will be detrimental to your conversion success.
It’s essential to gather data through A/B testing, discover patterns and trends, and analyze your visitors to convert these results into actions. When performing CRO and building a strategy, make sure to clearly outline the research process and make sure your team sticks to it.
The research process can (and probably will) change as time goes on, but finding your best CRO strategy early on will make sure you and your team always make the best decisions.
5. You don’t have the right tracking tools
CRO helps you measure and analyze website data. Without the proper tools, you and your team can’t accurately understand the data that you’re collecting and apply it to increase your conversion rate. You need to select the tools you and your team need for the process and include them in your conversion strategy.
Using a web analytics tool is a good place to start. Look for several features, including:
- A heat map
- Visitor recordings
- Geo-location insights
You’ll need statistics tools to help you calculate conversion rates and measure the success of your tested elements as well.
Here’s an example of using a heatmap tool to see where users are clicking most:
Conversion rate optimization is not an easy process. It takes time, money, and effort to get results and apply them to every test. By avoiding the mistakes highlighted above, you’ll be able to take a systematic approach to your tests and build a superior conversion strategy.
Guest author: Kurt Philip is the founder and CEO of Convertica, a done-for-you conversion rate optimization agency. You can get the Convertica team to audit your site for FREE. Visit convertica.org for details