This blog post is contributed by Janet Rausa Fuller. Janet is the content marketing manager at PickFu, a consumer research platform that provides in-depth feedback from real people on your creative ideas within minutes.
Every day, hundreds of millions of products compete for people’s attention and dollars on Amazon, Walmart, and other hugely popular marketplaces. And every day, both new and experienced e-commerce sellers must ask themselves this question: how can I stand out in such a hyper-competitive market?
With so many moving parts to running an e-commerce business, it’s not enough to offer what you think are the best products at a good price. Nor is it that straightforward: there’s keyword research to conduct, and ranking algorithms and seller guidelines to navigate.
But if there is one thing every seller should incorporate into their business strategy, it’s A/B testing. In this article, we’ll discuss what A/B testing is and how e-commerce sellers can reap its benefits.
What is A/B testing?
A/B testing, also called split testing, is an experiment comparing two versions of something to see which one performs better with users. In e-commerce, this means testing different versions and elements of your product listing or online store with customers.
A traditional A/B test splits traffic to your listing or site, directing some visitors to the “A” version and others to the “B” version, and tracks metrics such as page sessions and conversion rate.
The goal of A/B testing is to gather data on how your customers behave with each variant over a certain time period. Using this information, you can learn more about your audience and zero in on what to change or improve in your content in order to convert more customers.
How do you run an A/B test?
A/B testing of your e-commerce site can be done using landing page software such as Optimizely. You can run A/B tests via pay-per-click (PPC) advertising on platforms such as Facebook Ads and Google Ads.
If you sell on Amazon, you can also conduct A/B testing of your product listing or site within the marketplace, although it works a little differently. Rather than directing visitors to different versions of your listing simultaneously, an Amazon split test rotates the variants for a certain period of time and then compares the results.
For instance, you would test Option A of your Amazon listing for two weeks, then update it with different images or copy and see how that version, Option B, performs over a subsequent two-week period. There are a host of Amazon A/B testing tools that will do this work for you, including Splitly, Listing Dojo, and CashCowPro. (Amazon’s Manage Your Experiments feature, which is only available to select professional sellers, allows for testing of A+ Content and main images.)
Traditional A/B testing has its limitations. It’s usually a weeks-long process, and you need a healthy amount of traffic from the get-go in order to gather enough useful data. If you’re using PPC ads to drive traffic, the costs can add up. And because A/B testing happens live with actual customers, you run the risk of losing sales if one version of your listing doesn’t perform well.
One non-traditional route to A/B testing is via an instant polling software called PickFu. It runs independently of e-commerce marketplaces, which means you can test and get feedback before you launch. That way you’re not affecting your live listing, search ranking, or sales.
This software enables you to isolate and more quickly test a broader range of elements with a target audience of likely buyers, from your product title, description, and photos to the product concept itself. Results aren’t just quantitative; you also get qualitative feedback in the form of detailed written comments from respondents.
What are the benefits of A/B testing?
Sellers often make decisions about their e-commerce business based on intuition or what they think their customers want.
Francis Teo of Blue Lambda discusses the insights gleaned through A/B testing, “you will learn about what your customers want, and what gets them to buy. What appeals to them, and what does not. Once you have that learning, you can apply it to your store, in all parts of your conversion process. The results can be significant.”
With A/B testing, you collect real, quantifiable data on what’s getting clicks and engagement and what’s not. The more data you gather, the clearer that picture becomes. When you’re able to pinpoint problem areas in your listing or, conversely, see what kind of content resonates with more customers, you can feel confident that you’re taking the right steps to optimize your listing and drive sales.
You might split test two very different main images and see that shoppers prefer the one showing your product at an angle rather than head-on. You might experiment with shorter vs. longer product titles, or change the order of bullet points in your descriptions. Testing will tell you if these make a difference in audience behavior and sales.
Even better, with an instant polling software tool, you can gather detailed consumer feedback in the earliest stages of development, before you spend a dime on inventory or production. Not only do you save yourself precious resources now, you also put yourself in the best position when it’s time to launch your listing or e-commerce store, because you’ve already determined what resonates with your audience.
When should you use A/B testing?
There’s never a bad time to do A/B testing and especially for e-commerce sellers, there’s really no endpoint.
A/B testing is an incremental process. You should test continually and use the data from each round of testing to tweak your listings and online store accordingly.
Think of it as a data-driven marathon, not a one-time sprint. Thorough and constant testing helps you understand your customers better, improve their experience, and build a strong e-commerce business and brand.
A/B testing in action
Thrasio, a well-known FBA acquirer, stays nimble with A/B testing.
The company relied on multiple rounds of split testing when taking one of its brands, Angry Orange, a popular pet odor-eliminating concentrate, through a branding and packaging overhaul. The feedback not only informed the packaging redesign, it also led Thrasio to create an entirely new ready-to-use Angry Orange product.
The changes resulted in a 46% jump in sales volume within the first 20 days of updating the Amazon listings and a whopping 912% increase in daily revenue for the brand since Thrasio’s acquisition.
That’s an outsize example of the benefits of A/B testing. But small- and medium-sized e-commerce sellers report success stories too.
YES Bar, an emerging snack bar brand that sells its plant-based bars online and in some stores, was looking to boost its Amazon sales because conversions had hit a plateau.
The company, which began in the kitchen of a California mom whose son has food sensitivities, decided to focus on improving YES Bar’s main image on its Amazon product listing. The company created two new images that took up more white space and featured the snacks and packaging more prominently than the original. Then, it ran a PickFu poll to test which of the three images was most appealing to a target audience of 50 consumers ages 25-44. Respondents chose one of the two new images by a sizable margin.
Within two weeks after swapping out the old image with the new one, YES Bar saw a 11.8% jump in sessions, which the company projected would account for nearly $3,400 in increased revenue.
The bottom line
From customer service to listing optimization, a tremendous amount of thought and effort goes into running an e-commerce business and competing with the millions of other sellers out there.
A/B testing is one of the most effective ways to keep the momentum you’ve created going and claim your piece of the e-commerce market.