Less is More? How to Use Existence Testing to Improve eCommerce Sites
by Trevor Anderson • May 4, 2017
Want to get more sales from your eCommerce website? If you’re like most eCommerce businesses, you run your site a lot like a conventional retail store: the more options your customers have, the better.
After all, more options means your potential customers are more likely to find something they’d like to buy, right?
In theory, this idea makes sense, but conventional wisdom is not always data-driven wisdom. The “more options equals more sales” hypothesis doesn’t always pan out for eCommerce businesses.
Unlike shoppers in a conventional brick-and-mortar store, eCommerce shoppers aren’t forced to either buy what you have in stock or leave empty handed. A near-infinite number of alternative stores and products are just a click away. You’re competing with the internet, not a 10 minute drive to Wal-Mart.
eCommerce is the epitomy of window-shopping, so if you want people to buy, you need to make them feel immediately connected to your product. In many cases, that means less options and potential distractions—not more.
Less is More
Existence testing is the art of strategically removing website elements to see how those elements affect purchase behavior. In essence, existence testing is all about asking the question, “What happens if I get rid of X?”
Existence testing is based on the “paradox of choice” marketing theory. Put simply, the paradox of choice states that when users are given too many options, they will tend to choose nothing at all.
Researchers have studied this phenomenon for years, but the most famous study actually took place in a grocery store. In this study, researchers put up two different types of jam displays: one with 6 types of jam and one with 24 types of jam.
As it turned out, people who visited the 24-jam display had one-tenth the purchase rate of people who visited the 6-jam display. The 24-jam display drew in a lot more people, but produced a much lower purchase rate.
Why? Option overload.
With 24 options to choose from, people couldn’t decide what they wanted and ultimately decided not to make a purchase.
Existence Testing in Action
To show you how this works, we recently ran a series of existence tests on a client’s large scale eCommerce site. This client sells 371 products that fit into about 12 different categories.
Shortly before we started working with the client, they had redesigned their home page. Putting their heads together, the marketing team had come up with a collection of products that with high margins and/or high sales volume and decided that would be a good lineup to place on the home page.
Many companies take this same approach to their site design. It’s not a bad approach, but it can cause problems if you don’t take the time to test it.
So, that’s what we did. Since the page was new and unproven, our approach was simple. We theorized that many of the products on the page were not the perfect fit for a consumer, and the overload of possibilities decreased the likelihood of purchase.
In the initial round of testing, we created 10 variants, with each variant removing one element on the page. So, as a whole, we tested the entire page…removing one element at a time.
Here’s it works. In this test, we removed the beard care section from the page (this is a mobile version of the client’s home page):
Astonishingly, this test revealed six winners and produced a combined lift (between all of the variants) of thousands in revenue in 2 weeks. The top 3 variations improved conversion rates by margins of +9.4%, +18.07%, and +10.9% over the baseline. Talk about a big win from a small test!
What happened? Well, as in the jam study, less options led to more sales. When the client’s users were presented with fewer options, they were more likely to find something they felt strongly enough about to purchase. While some items on the site “tasted good”, they did not taste as good when it came time to spend money.
While this was just the beginning of testing on the home page for this site, it gave us a great starting place for our later tests (stay tuned for articles on multivariate and ordering testing).
If, like most eCommerce marketers, you probably feel the obligation to maximize options on your site. However, before you add more options, ask yourself, “Are all those products really necessary?”
More often than you might think, the answer may be “no.” But, the only real way to answer that question is to use existence testing to see which site elements really affect your purchase rate.
If you are willing to risk your pride and challenge the intuition of your marketing team, existence testing is the way to go. While it may be a bit painful to second-guess your site design, what you learn about your core audience and the direct returns are well worth the investment.
By the way, if you’d like me to take a look at your site and suggest some existence testing ideas, let me know here or in the comments. I’d love to help!
Have any questions or experiences with existence testing? We’d love to hear from you in the comments below about your experiences with existence testing and how it has helped your site and/or helped you learn new things about your customers.