Harness The Power of AB Testing Tools: The Digital Marketers Superpower

andrew browne Split Testing Best Practices Leave a Comment

Sharing is caring!

Online marketers have been using AB testing to help make changes to their websites for years.


By split testing, online marketers have been able to scientifically optimize websites and other ad channels to meet almost any goal.


However, until now, Amazon private label sellers had no way of AB testing their private label listings. Now Splitly, which is our AB testing tool, allows Amazon Private label sellers to split test their Amazon listings.


Private label sellers can now use the power of AB testing to increase sessions, conversion rates, sales and profit.


Which Version Is Better? Only Split Testing Can Tell You


Online marketers use AB Testing (also called split testing) to compare two versions of a web page to see which performs better based on their end goals.


AB testing tools split your visitors into different groups and shows them different versions of your website. This is usually done at the same time to control for differences that time may be cause, including fluctuations in traffic due to changes in the day of the week.


Usually when you are AB testing, you are testing two variables – the control (original website with no change), and the other version of the website with a slight change.  These two variables are usually called A and B. Hence the name AB testing. Once the test is done, you compare the data from one variable against the data from the other variable.


However, you could test more than two variables. In order to test more than two variables, you should have as many groups as you do variables because each group will see a different variable.


Therefore, if you are testing three variables you should have 3 groups. Each group would view three different versions of your website.


However, there is a test where a group sees more than one variable called a multivariate test.


You Can Test More Than One Variable at a Time Using a Multivariate Test:  


Multivariate testing uses the same basic premises of AB testing, except that in this test, you are comparing a higher number of variables and testing them together. Multivariate tests help reveal more information about how two variables interact with one another 


The real power of a multivariate test is to test different design combinations to see which one has higher conversion rates. It can also reveal which elements have the biggest positive or negative impact.


The only major drawback of a multivariate test is that there can be a lot of different combinations to test, which means that you will need a larger audience to view all these combinations. If your website does not have enough traffic, you will not be able to run a multivariate test.


Meet Your Marketing Goals With Split Tests


You may be asking yourself, “why would someone go through all this trouble?” The truth is, AB testing is a powerful tool.


By comparing two elements of a website, while making sure that other variants are not affecting the results of the test, you are able to see the most effective methods of improving your website.


If you conduct several tests, and find several variables that improve your website, you can combine these variables into a very well optimized site.


This is especially powerful when trying to increase conversions in existing traffic.


For example, you may already have 1,000 visitors a month to your website. Let’s say out of that 1,000, 10% convert at $100 each. If you were able to split test a part of this website, let’s say a “call to action” button, and increase your conversions to 20%, you just made an extra $10,000 without doing anything to increase your traffic.


Split Tests: Real World Examples


Optimizely gives great real world examples of how companies used AB testing to increase conversions, unique page views, engagement, and so forth.


For examples, Sony was able to increase purchases by 20% by AB testing variations of its checkout page.


ComScore was able to increase leads by 69% by AB split testing their landing page.


Secret Escapes AB tested their mobile signup page and doubled their conversion rates.


Unbounce also notes many beneficial effects of successful AB split tests.


For example, Unbounce was able to increase signups by 24% by AB testing images in their signup sheet.


They were also able to increase leads by 368.5% by AB split testing different sign up forms.


AB Tests & Amazon Private Label Sellers


Do you have a private label product? Do you constantly find yourself trying to decide:


  • What is the best sales price
  • What is the best listing price
  • Which pictures are the best to use
  • Which title is the best to use
  • Which bullet points should you put into your listing
  • How to describe your product
  • How many keywords to use


What if you could harness the power of AB split testing to address these questions and compare any changes you make to your listing with the original listing? You would be able to find your most affective changes, and implement them permanently on your listing.


Splitly allows you to harness this power by helping you conduct AB tests on your listings.


AB Test To Optimize Your Private Label


We already discussed how AB testing can be used by online marketers to make powerfly optimized websites. Now, private label sellers can use Splitly to conduct such tests as well.


Most Amazon private label sellers randomly change things in their listings thinking these changes will improve their listing. However, there is no science behind these changes.


These changes are based on random ideas and then haphazardly thrown together in a listing by the owner with hopes they will increase sales.


These owners are making changes in their listing based a hunches. Most of the time they keep these changes simply because they “look better”.


Don’t make the same mistake. Any change you make on your listing should be made based on data focused on optimizing your listings.


The online marketer does not just make changes to their website based on hunches. Quite the opposite. They improve their websites through split testing.


The online marketer uses split testing to methodically test several variants and then uses data to guide their decisions. There is no guessing in their work. It’s all based on facts.


You must do the same for your Amazon private label listings.


Automate Split Tests With Splitly


Unlike AB testing for websites, which had a number of tools to help you split test, AB testing for Amazon private label listings had no tools.


Before, if you wanted to do a split test for an Amazon listing, you had to

  • Manually go in and make the changes
  • track the changes
  • track the time you need to switch in and out the variables
  • Manually switch in and out the variables
  • hope that all other variants were controlled for
  • collect all relevant data on your own including sessions, sales, conversion rate and profit
  • analyze the data on your own
  • calculate statistical significance
  • guess whether any change in data was due to the changes you made or random fluctuations


As you can see, this method is not only a lot of work, but it is not accurate as you have to guess whether the changes you made caused the fluctuations in data.


However, Splitly will now do this work for you. It also tracks and presents all the data in an easy to read display. Splitly also calculates statistical significance to help you determine whether any changes in your data was caused by your variables or random fluctuations.


Splitly Lets You Test (Virtually) Anything


Splitly allows you to AB test almost everything in your Amazon listing.


outcomeiconsArtboard 3

For example you can test

  • Your advertised price
  • Your sales price
  • Your titles
  • Your photos
  • Your bullet points
  • Your descriptions


As you can see, it allows you to test most variables in your listings.


You can also optimize your listing with several goals in mind including conversion rate, average daily sales, average daily sessions, and average daily profits.


Quick note: When using Splitly, we recommend that you don’t use a multivariate tests unless you really know what you are doing. They take much longer to get a result, and if setup incorrectly, the result will be misleading or incorrect.


In Conclusion


AB split testing has been a scientific way to fully optimize website by testing different variables and seeing which ones are better based on the testers’ end goals.


By AB Split testing, online marketers have been able to scientifically optimize websites.


Now Splitly, which is our AB split testing tool, allows Amazon private label sellers to split test their Amazon listings.


This brings the power of AB split testing to Amazon sellers who can now use it to optimize their listings.

Sharing is caring!

Leave a Reply

Your email address will not be published. Required fields are marked *