A/B Testing

The page that does not convert is a non-value page for your business. We offer scientific solutions with A/B testing services to meet your macro and micro conversion goals.

If you want your visitors to convert more to improve your profits; our conversion rate optimization work will meet your needs. With the A/B test, we offer within our CRO work, we produce variations that can increase the conversion rates of the elements in pages with traffic (title, visual, Font, CTA (call to action), etc.), creating these variations in different pages. and putting them in the race. We increase your conversion rates by detecting the best performing variation and applying it to your page. Contact us to ask questions about CRO and A/B tests and find out about solutions that we can produce exclusively for you.

Illustration about A/B testing

A/B Testing Process

Below you will find detailed information about all of the work we do during the A / B testing process. To summarize this process, we first install the softwares that generates heat maps of web pages. When the heat maps have enough data (this time varies depending on to the traffic the page receives), we examine the generated heat maps data and Google Analytics data (or any other analytics tools you use). As a result of our review, we are detecting problems that may be preventing your website visitors from converting and we create solutions that can encourage visitors to convert. We subject these solutions to single-variant tests (A/B testing).

The single variation, i.e. A/B test, is the competition of the performance of different variations of a particular element. For example, when we think that the title you use should be improved, we create pages that contain different variations of this title. Visitors are equally guided to these pages with different variations and their behaviour is followed. When enough data is reached, we determine which header usage positively affects your conversion rates and this header is placing on the page. A/B tests that maintained through different elements continue to improve the page’s conversion rate performance.

1. Predicting Problems and Possible Fixes

In order to identify possible problems and solutions that can be produced for these problems, we first install software that will allow the creation of a heat map (if there is not) to your CRO-work page. While data is collected for the heat map, we are adding conversion goals to your Google Analytics account. In this way, conversion rates are becoming measurable through Google Analytics. When enough data is collected in both your Google Analytics account and the heat map (the ideal value is a thousand visitors, but the ideal value may differ based on the traffic of the page) we analyse this data with our expert team. As a result of our analyses, we are producing possible solutions for these problems by identifying problems that may pose an obstacle to the realization of the visitors ‘ conversion actions. We evaluate all possibilities by taking your feedback at the stage of the solution generation.

2. Running an A/B Test

Instead of directly applying the solutions we produce, we undergo A/B testing. Because it is almost impossible to predict the behavior of visitors by 100%. Unexpected changes may perform better. We are testing our solution ideas with A/B test to ensure that our CRO works are based on statistics and analytics, not based on forecasting and thoughts. During the A/B test, we create new pages that contain different variations of the change to only one element at a time. We redirect an equal number of visitors to new pages created. In this way, we have data that we can assess each one’s performance in separate and equal conditions. We also set up heat map software and Google Analytics conversion goals on pages with different variations, such as on a CRO page. According to page traffic, we are waiting until we have enough data.

3. Analyzing Results

We begin analyzing this data when there is enough data to be analyzed on heat maps and on Google Analytics. We examine the pages competing in the A/B test according to both macro and micro conversion rates. We give points to macro and micro transformations to detect the variation that shows the best performance. For example, if the visitor spends at least 30 seconds on the page, it is 10 points, if they fill out the contact form and send, it is 200 points… In this way, we incorporate all conversions of the page into a performance analysis without having to be plugged into a single conversion target. With the average macro and micro conversion counts occurring on the pages, we multiply the scores of these coversions to calculate the total conversion score of the page. We do this for all of the pages in the A/B test to determine the best performing variation in the conversion rate. By comparing the performance of this variation with the performance of the original element, we calculate the performance increase that can be obtained in CRO.

4. Implemention and Repeat

Immediately after determining the variation with A/B test, which has a better conversion performance compared to the other variations of the modified element and the original, we use the variation set on the CRO page. We then continue to regularly monitor the performance of the modified item. However, our A/B tests do not end with the improvement of a single element. We repeat this process for all items that may pose a problem that we have identified at the first stage. In this way, until each element of your page is encouraging for visitors to take conversion actions, our tests are ongoing, and your conversion rate can result in noticeable results. If your CRO page has a high traffic volume, we will perform multiple A/B tests at the same time to accelerate this process. Contact us for A/B test and to inquire about the questions you have in mind about CRO work and about the solutions we can offer you.

Difference of Multivariate Testing

If you have done research on CRO, another terminology term you will encounter is multivariate tests. The ability to distinguish multivariate tests from A/B tests is to test multiple element variations at the same time. For example, when comparing the performance of different page titles with A/B test; in multivariate tests, variations of multiple elements, such as the page title and the banner image, are compared. Although this approach accelerates the CRO process, there are also negative aspects. One of these negative aspects is the inability to make sure which of the modified elements provides a performance increase. As in our previous example, it is not possible to determine exactly which element received the performance increase compared to both page title and banner image variations at the same time. While an element variation improves the conversion rate, the other element variation may be a bad influence, and it is very difficult to detect. Which type of test to use varies depending on the page and the analysis of the page. As Growth Marketing, we are implementing the most appropriate CRO methods that can be applied on your website.

tristan_empatika

These guys are fantastic! Extremely professional, and delivered some exceptional results.

Tristan
Founder, Empatika
josh_seo-guru

I’ve been in SEO and digital marketing more than 10 years. There are few people I trust on these subjects and Ya┼čar (Founder of GrowthMar) is one of them.

Josh
SEO Guru

My website had been stucked in position 3-4 on Google for so long. Thanks to GrowthMar, now it’s #1!

Gary
CEO, JagBits

Get in Touch

We would be more than happy to talk to you about our services. You can contact us via the form below or send us an e-mail at contact@growthmar.com!