Conversion Rate Optimization Myths


Myth 1: Conversion rate optimization is all about A/B testing like changing button colors, text and call-to-action buttons etc

There is no denying the fact that tweaking your colors, texts and call-to-action buttons might result in slightly higher conversion rates, but it is not the ultimate solution to low conversion rates. You might hear someone saying something like: “I increased my conversion rate by two percent when I changed the CTA button color from green to red.”

It is meaningless, however, to try to increase your conversion rate if you don’t know the reason behind the increase in conversion rate. Have you studied the purchasing behavior of your consumers? You can only speak with authority once you know the reason for the change.

There are more serious barriers to conversion optimization that you should be worried of. Each of these barriers have solutions that almost certainly does not require changing the colors of your buttons. The best way to overcome barriers to conversion is by rolling up your sleeves and putting yourself in your customers’ shoes.

There are many possible reasons you are not converting:

  • Your website is slow;
  • You have not earned your visitors’ trust;
  • Your copy is confusing;
  • A lengthy conversion process that is both discouraging and confusing;
  • A complicated CTA button that is obstructed by unnecessary texts, links, images and copy;
  • Your landing page text/offer does not tally with your ad text;
  • Your pricing is outrageous;
  • You don’t offer a differentiating factor.

Myth 2: Conversion rate optimization can only be done by professionals

It is quite obvious that you need professionals to do your conversion optimization because of the expertise, tools and experience they have gained over the years. However, that doesn’t prevent you from digging in your analytics to find out what is wrong and ask yourself ‘why?’ You don’t need a professional to do everything for you. You need to understand the basics of conversion rate optimization so that you can do your part.

Myth 3: Statistical significant means revenue

When you’re finished running a test using Optimizely, Visual Website Optimizer, or any other A/B testing platform, you will receive a summary of results.

Let’s say the results show you nearly doubled the conversion rate using your winning variable at a statistical significance of 95 percent. Does this then mean that you are going to see twice as much revenue? The simple answer is NO. Although you might see a slight improvement in revenue, it is possible that you might not see the same percentage in revenue increase for various reasons:

  • Traffic sources change: Sometimes website traffic fluctuates for various reasons. Maybe a celebrity decided to tweet or share on Pinterest some of your content causing a spike in your traffic — this means there is no guarantee that you will maintain that traffic. You can have very misleading results even at 99 percent confidence interval.
  • Fluctuation in website performance: Your website performance might fluctuate. Did you know that slow speed negatively impacts your conversion rates? Now you know. If at one time your website performance became better or worse for whatever reason, you might see an increase or a decrease in your conversion rate. Unless you are going to make a conscious effort to maintain the consistency of your website performance, you can’t be certain that your recent increase in conversion will automatically lead to an increase in revenue.
  • It could be insignificant: We often take for granted that if we have 95 percent of 99 percent confidence interval, that we have definitively beaten the baseline conversion rate by 95 percent or 99 percent. What about the five percent or one percent? We forget that we could still be part of the unlucky five percent or one percent.

As a general recommendation, try to conduct your research over an extended period of time, say for one full year, to ensure you are actually seeing stronger results rather than relying on statistical significance alone. 

Myth 4: always be testing

Conversion optimization often involves testing using A/B or multivariate testing. I cannot overemphasize the importance of carrying out these tests because they will help you figure out what works and what doesn’t work in your website with headlines, design, site copy, button colors and text, etc. Testing, however, is only one of the steps in conversion optimization. In other words, you don’t just wake up one morning and starting running tests on your site elements. There are a zillion of elements that can be tested; so where should you begin?

Often, conversion optimization involves the following steps:

  • Collecting data and analyzing it;
  • Determining the problem and which pages should be improved;
  • Prioritizing your problems in order to start with the most significant one;
  • Drafting a hypothesis on how to fix the problem;
  • Implementing the hypothetical solution on the website;
  • Running A/B test (or multivariate test) to determine the winning design;
  • Studying the result of the test;
  • Starting from step 4 with your next problem and repeating the steps.

As you can see, testing is only one step in the whole process.

Myth 5: You can guess your customer needs

Of all the myths in CRO, this could well be the most absurd. Many marketers think they know their users well, hence no need for surveys and testing. If that is what you have been thinking, you are in for a big surprise.

If you think your users love short content because they have neither the time nor the attention span to read long content, you are probably be wrong. You could mapping your own set of beliefs and preferences on your users or you are simply implementing what you read on the Internet. For instance, if you read somewhere that red ‘Buy Now’ buttons convert better, you might go home thinking, “I should also change my ‘Buy Now’ button to red.

While those assumptions can form an important process in hypothesis formation, there is no way of knowing if they are correct unless you have empirically tested them and studied the results. Certainly, you can’t overlook the importance of surveys and polls—raw customer feedback.