Unfortunately, there are no 3rd party tools that perform true A/B testing on Amazon, since we can’t have duplicate listings. True A/B tests need to run simultaneously so times and dates don’t affect performance.
But there are some Amazon seller tools that perform Before and After tests, like SellZone.
Example: Listing A with the exact same bullet points and images and Listing B with a different tile, both run at the same time and after a period of time we compare the conversion rate, click-through rate, etc.
With that said, every single change I do gets tested. This way I can get real results, instead of just believing someone’s guess on FB.
A/B testing is a technique used to compare two versions of a web page in order to find out which one performs better. It is often used in marketing and conversion optimization, but it can be applied to any situation where the goal is to increase the conversion rate.
The most popular type of A/B test is the split test, which divides visitors into two groups, showing different versions of a web page. The conversion rate for each group is then analyzed and compared.
In this article, we will explore what an A/B test is and why it’s so important for your Amazon marketing strategy.
A/B testing can be a powerful tool for improving the performance of digital marketing campaigns, and of course our Amazon PPC campaigns. It allows marketers to test different versions of a listing and measure which one performs better in order to determine how to improve future pages or other aspects of their campaign.
There are three components to A/B testing:
The original version, the new version, and an element that will be changed between them.
One cannot simply update the title or the product description and then expect to see an increase in sales. Everything has to be measured, analyzed, and researched so that you can have accurate statistical data.
Then you can use this data to improve your listing’s conversion, which will then increase your organic ranking on Amazon’s search engine.
When it comes to listing, there are many things that you should split test.
For example, testing whether the price affects the conversion rate or not. You might want to compare what happens when your product is priced at $10 or $20. Or you could test what happens when you show a video of your product versus not showing a video of your product.
There are many other things that you can split test on your listings too, so it’s important to be creative and think about what might work for your specific business.
You can trust Amazon sellers to understand that it doesn’t work for any specific business model you need. If your competitor sells exactly the same product as your own, they can probably take the profits and grab your traffic. Developing an effective and fully optimized product detail page will increase the conversion rate (Product sales) and can also increase customer loyalty.
It is crucial to run split tests on different parts of a listing, not only on the product title, or product images. You can do so with a split-testing tool, or even better if you are a brand-registered seller you can run a new experiment inside the seller central account.
Split tests use the concept of choosing a website element to optimize and running a combination of 2 / 4 versions simultaneously for analysis of the customer response. Split Amazon is different if you don’t have the Amazon brand registry. Unless the user has not registered, you must start the site in succession for a specific timeframe. However, the registration in the brand registry provides retailers with an online version control system for analyzing and evaluating product versions of their listings that allow continuous listing improvements over time.
Every Amazon seller knows how valuable data can be, but not all brand owners use data to better their conversion rates. I often see product titles that are based on simple keyword research as if that was a seller’s secret weapon, but they are actually put together as that seller “Assumes” it will perform the best. This often results in fewer sales and of course smaller conversion rates.
So remember, do not assume the best image is the one you like the most, or the best title is the one that has ALL the keywords you found on a KW research. Customer behavior and listing metrics (real data) is the thing that will get you more sales.
Not all Amazon sellers have the opportunity to use the management tool. These are sellers registered with Amazon with high-traffic ASIN numbers. You need traffic or sales because your analysis needs statistical information about your listing and customers. You will also notice that managing your experiment will display if you can use an asin for your portfolio. Use Amazon ads to increase the number of users when you can’t use the Manage Your Experiments tools.
You can test:
Product Title
Product images (Main image)
A+ Content
Bullet points
Product Description
If someone tells you that Amazon A B testing needs to run for at least two weeks, stop listening!
The duration of the test is not important, but what matters is the size of your sample test to give you enough data or statistical evidence of the customer’s choices.
For example, split testing performed for at least two weeks with a product that gets 10 impressions per day, will have less evidence of units sold and unit session percentage, than high traffic asins that gets 10,000 impressions per day and the split test example only runs 2 days!
In short, if you are running a pricing strategy, split testing using the 3rd party split testing tool, make sure that tool gets as much data as possible.
One of the first tools I ever heard that would perform a split test on Amazon was called Splitly Amazon ab testing, and then Sellerly, which became Sellzone.
These tools performed what is called, before and after split tests. This means they did not run the two versions of the test simultaneously, but rather they would run one test for a week or so, then run the other test.
This is not the best way to perform an AB test, because there are several things that can interfere with the tests during the 2 or 4-week period you do them.
Imagine running a test for 1 week in July, and the second test is performed during the 1st week of August when most people are on Vacation. Or even some perform 3-day tests and the second test captures the weekend.
In some cases, these types of tests can actually hurt your optimization if you don’t have a big enough data sample to prove version A or B is the winner.
Although this post talks about Amazon sellers and seller central account split testing, it doesn’t mean it is only for Amazon a b testing. This can and should be used in all e-commerce platforms and all marketplaces.
Soon we should write about A/B testing while selling on Walmart
Make sure to read our Amazon Consultant article