After going through my recent article that covered the complete guide about the A/B testing technique, someone asked me about why I have not discussed about the tools used for implementing the A/B test accurately on the website or if this test will have any impact on the overall SEO of a website? In response to his query, I am going to shed some lights on few key tools that are used for providing simplified A/B testing functionality as an important part of the Google Analytics suite. Besides, I will also discuss about the impact of A/B testing on the SEO of a website.
If you aren’t aware of the basics of A/B testing, I will recommend you to go through my first article to brush up your knowledge about its fundamentals and why it is so important for all marketers and designers? Now, here is the brief information on the best three A/B testing tools, before moving forward with procedure of setting up A/B test with Google Analytics and its affect on SEO.
Google website optimizer is a free A/B testing tool that is used for creating experiments for testing of conversion rate. For using this tool, you have to establish content experiments for different elements on your web pages and then Google will randomly serve the varied versions of your web pages to the real number of visitors who come to your website.
You can easily test up to 10 different variations of a single landing page. With Google website optimizer, you already know that the participants in the tests are your target audiences and hence, this tool has great advantage over other A/B testing tools when it comes to focusing on your target market.
Tips for using the full potential of Google website optimizer:
Visual website optimizer is an effective A/B testing tool that offers you the easiest and quickest way for increasing your website leads and sales. As a leading optimization and testing tool, it allows the product managers, marketers, and analysts to create A/B tests without having any proficiency in HTML or technical knowledge. With this tool, the marketing professionals can easily create different variations of their landing pages and websites and then with the help of a check point editor, they could check which of these versions are able to generate maximum leads and sales. This tool also allows its easy integration with the Google Analytics and hence, you can conveniently perform the analytics of your A/B test in Google Analytics.
Benefits of Visual Website Optimizer:
An incredible A/B testing tool, Unbounce helps you in making impressive landing pages and testing new ideas and hypothesis. This tool claims to increase conversion rates up to 20% by implementing new design changes and ideas for A/B testing. It comes with effectual drag and drop landing page builder that assists you in creating responsive web pages without seeking any assistance from the technical team. The “code-free zone” of Unbounce tool creates and publishes a high conversion landing page in a single click.
Benefits of using Unbounce tool:
Once, these 3 important A/B testing tools have been compared, the next step is to set up the A/B test in Google Analytics and for this we will organize the test using Google website Optimizer tool (also known as Google Content Experiments).
Google Content experiment is the advanced version of Google website optimizer tool that is used for running A/B test from inside Google Analytics. For creating the test with this tool, there are some important steps that have to be followed and they are listed as follows:
For creation of content experiment, you have to navigate the Behavior section of Google Analytics, where you will find “Experiments” link on the sidebar. If this is a completely new and first experiment for you, then you can begin by clicking on START EXPERIMENTING, however, if you have already existing experiments displayed on the age, then click on the Create Experiments button and a new window will open asking for the given fields:
When you have selected the experiment and finalized the experiment objective, click on Next Step. A New window will open in front of you like this:
You can add the variations and the URL of the webpage that you will like to test and you will see the thumbnails of the web page to ensure that you are entering the right URLs. When you have filled the required details, you can click on “Next” button to proceed with the subsequent step of setting up of A/B testing.
For setting up the content experiment code, you have to select either the required code for running your test or you will be given with an option to send an email to concern person who is implementing the code. You can also receive the Experiment ID that will be used in the subsequent step for implementing the experiment. You can get the Experiment ID by clicking on the option “Manually insert the code” button.
Once, the experiment ID is received, you can move forward with the next step of validating and confirming the experiment code.
This step confirms the substantiation of the content experiment code and if the code is missing, an error message will be displayed. However, you can skip this step if you want to do so, and for this, click on “Start Experiment. A pop up message comes up with the following information, “Experiment validation had errors or did not complete. Are you sure you want to start the experiment? If you are sure that your experiment is properly set up, you may continue.” However, it is advised to check the code for evaluating why the error is displayed and then you can try the substantiation once again
Once the content experiments code is substantiated and you have configured it appropriately, your next vital step will be to run and publish the experiments depending on the results. With its proper execution, the users will be included in the experiments and this will help you in understanding which of these variations are receiving maximum traffic without any redirects. Once, the experiment is executed and you have come across the results, you can easily include the changes such as CSS and HTML of the original page and remove the experiments.
Now, when you have learned about the tools and implementation of A/B testing within Google Analytics, the consequent query that might pop up in your mind is that, “Does the A/B testing impact the SEO of your website”?
Here are some guidelines of Google that all professional marketers should follow to avoid penalties and ranking affect on their website during A/B testing.
Google has strictly posted the guidelines to the websites performing A/B testing that they should avoid cloaking and display the crawlers what type of changes are happening in your website. Ensure that you can never decide whether you have to serve the test or the content variant depending on user-agent. Always serve the original content to the user-agent and breaching with the guidelines can make your website removed or demoted from the search results of Google.
Google recommends the marketers to use rel=”canonical” method to make sure that the variations of their web pages should be considered by the search engines as closely related with the original URL because if you don’t use the rel=”canonical” then, it could impact the ranking of your web pages in a negative way, which is perhaps not what you are expecting to come out as the testing results.
It is wise to use the temporary redirect “302” instead of “301” because it helps the search engines to recognize that the redirect is temporary and it will be there only till the time you are performing the A/B testing on your web page. Hence, they will maintain the ranking of your original web page instead of changing it.
Running experiments for longer duration can initialize penalty from Google. It is necessary to update your website once the test is concluded with the selected variation. Running experiments for longer than Google expectations will be treated as an effort to mislead search engines and hence, it might raise penalty on your site.
When you will follow these guidelines of Google while executing A/B testing on your website, it promises not to affect the SEO of your site, which in turn will maintain your site’s rank in search results.
It is possible to test more than one thing at a time, but then it will become multivariate testing instead of A/B testing, which will also take unusually longer durations. For example, let’s say you want to test three different variations of a particular call-to-action button on your web page. In such cases, if you run a single test by splitting the target visitors on these three variations equally, you will be able to run a more efficient A/B test.
Besides, running more than one thing at a time is also more complicated as it requires more days to run and check the results accurately.
Hopefully, you must have been inspired to run the A/B testing appropriately on your website by going through this post. If you want to share your own thoughts on A/B testing, please write your comments in the section given below.