skip to Main Content
A/B Testing Ideas For Ecommerce Stores

A/B Testing Ideas for Ecommerce Stores

Tis the season for ecommerce stores. We?re now well into the busiest time of year and the biggest opportunity for store owners to make bank.

This may also be a good time for you to run some A/B tests on your ecommerce store. While there are no set tests that work for every ecommerce store, there are some testing ideas that seem to get better win rates than others.

Of course, there?s no guarantee for what will work on your particular store – testing involves many variables, including your specific audience. As always, we suggest that before you embark on any tests, you gather some solid baseline data on your website and have some clear goals for what you?re hoping to achieve first.

Here are a few ideas that have worked for ecommerce stores:

Get our checklist of 20 situations for ecommerce testing here

Your shipping offer

Around 75 percent of online shopping carts are abandoned by shoppers, accounting for a massive number of lost sales. Of course, there are a variety of reasons why a customer might ditch their cart, including factors that you can?t really help, such as the tendency to add things to a cart while shopping around.

For the factors that you can do something about, shipping is an issue that ranks highly. A Baymard study found that extra costs that were perceived to be too high (shipping, tax or fees) was a factor for 60 percent of those surveyed abandoning the cart. In a world where Amazon has normalized free and fast shipping, other ecommerce stores are being hit with increased expectations from shoppers.

One of the major concerns of smaller ecommerce stores is that shipping is expensive. Offering free shipping might help with conversions, but it also may be a profitability killer. The only way to find out is to experiment. Some suggestions include:

  1. Know your baseline. What is your conversion rate right now without a free shipping offer? What happens when you do offer free shipping? Is there a sufficient increase in conversions to buffer the cost of shipping?
  2. What happens if you cover all or part of the cost of shipping by increasing prices? Do you see improvement in conversions by offering free shipping?
  3. Try establishing a threshold above which the shopper earns free shipping, for example ?free shipping on all orders above ?40.? Analyse the impact of this – does average order value increase? Do you improve overall conversions?
A/B testing

The Kiwi Importer makes shipping options clear from the homepage

Make your contact information prominent

Trust is a big issue for any online shopper. If they can?t physically handle the goods and speak to a person, they want to know that they can safely place an order and get what they?re expecting.

To that end, many shoppers will have questions. They?ll also look for information about your store, including whether you can easily be contacted. ?I didn?t trust the site with my credit card information? was a reason given by 19 percent of respondents in the Baymard study, so adding elements to your site to promote that trust is important!

You?ll see a lot of sites now place information for how to contact them much more prominently on the page. If you haven?t already, this may be something you could experiment with. Your shoppers want to feel that they can easily get hold of you if something were to go wrong.

Ecommerce shoppers always look for clues that they can trust the website Click To Tweet

Make other ?trust? elements prominent

What else does a shopper need in order to find you trustworthy? According to the Baymard study, return policies were also a factor in cart abandonment. Either the policy itself was not satisfactory, or the shopper can?t immediately find it so they give up.

Other elements include things like your product warranties and/or any guarantees made about the product. The customer wants to know that they have recourse if the product doesn?t work as expected or is faulty.

You can test whether making your returns, warranties and guarantees policies more prominent helps with conversion rates versus the ?control? state you already had.

Static images vs. carousels

There is a lot of debate in the conversion optimisation world about static images versus carousels. In some cases, the carousel may work, but in most that have been properly tested, the carousel is a failure.

You only have to recall your own website experiences with them to understand why. You see something that might be interesting, only for it to flick past before you?ve had a chance to look properly. Trying to get back to what you wanted to see can be a usability nightmare, and anything else on the carousel loses visibility.

Studies have proven what we can understand about them intuitively – carousels irritate users and reduce visibility of the messaging. Overall, the user loses control of their interaction with the site and use of the carousel can create banner blindness.

A lot of ecommerce sites are still using carousels in the header area of their site, often in an attempt to get multiple messages across. If you are one of those sites, consider replacing the carousel with a static image that has your best value proposition at the time (for example, that might change to be a seasonal sale image at this time of year).

Emphasise best sellers on your homepage

You ideally want shoppers to go beyond just your homepage, and a good way to do this is to get them clicking on something of interest. Many ecommerce sites do this by highlighting their best sellers or top deals on the homepage so that people click through to the product.

There are also good arguments to be made for limiting the options you display on your homepage. Experiments have shown that too many may be overwhelming, lowering your conversion rate.

A now-famous experiment was conducted by two marketing professors. They offered jam tastings at a supermarket booth for two Saturdays – on the first one they had 24 flavors, the next they had just six.

Overall, 60 percent of customers stopped to taste the jam when there were 24 flavors, but only 40 percent stopped when there were six flavors. You might assume that meant that having 24 flavors was better, but in looking more closely at the results, of those who stopped for the 24 jams, 3 percent made a purchase, while 30 percent of those who stopped for the six jam flavors made a purchase.

They concluded that offering more products enticed more people to stop, but that doesn?t translate to purchases. The end goal is to get customers to purchase, so offering just a few of your best sellers may be better!

Test your product descriptions

Product descriptions should tend to evolve over time, especially as you learn more about your customers. The aim is to give them enough information that it answers key questions they would have about the product, without being a lengthy wall of text that no one wants to read.

That point is important – your descriptions should provide clarity rather than being a sales push. You may have clues lying in the questions customers have asked – what did they want to know before buying?

Test whether enhanced product descriptions beat your old descriptions through testing. Check that the description is the only difference though – sometimes other elements may play a role. In a test highlighted by Crazy Egg, a motocross clothing company tested out a comprehensive product description versus their old version – no description. Surprisingly, the control of no description won. On examining the page though, the new descriptions were arranged at the top, pushing the product image down. This may have played a role in the results.

A/B testing

Test progress indicators on checkout

Baymard?s survey on cart abandonment showed that many people felt that the checkout process was too long or complicated, hence they abandoned the cart. This can happen when the person is faced with a huge long page to complete, or they keep getting taken from step to step, in a seemingly endless process.

There are a couple of lessons here; a) keep your checkout process as simple as possible, and b) let people know how much progress they have made through the checkout!

People always want to know how long something is going to take, so some kind of progress indicator (such as a numbered list) can help with this. In fact, you can devise several different tests from this, depending on what you already have set up. You might alter the number of steps you have, how they are presented (single page vs. multi-page) or how progress is displayed.

Free download: Common situations for ecommerce A/B testing

Final thoughts

The ecommerce business is highly competitive, so all stores should be looking at A/B testing for conversion optimisation. With the biggest sale period of the year upon us, this is even more important.

The list of things you could test is virtually endless, but here we?ve looked at a few items that have commonly produced results for ecommerce stores. Importantly, rather than testing at random, be intentional about what you?d like to achieve and select areas that you feel are ripe for improvement.

If you need help with this, of course, you can come to a company like Team Croco! We set up and run experiments from start to finish for ecommerce stores, saving you a considerable amount of time.

Back To Top