How To Reduce Shopping Cart Abandonment With User Testing

Bonus Tip #1: Get Real

Before you test, make sure you remove all roadblocks to getting honest and complete feedback.

• Use REAL money (yours), or developer accounts. If you recruit a tester to buy, say, a washing machine, and the tester doesn’t actually feel like laying down $1000 during the test (go figure), you could severely limit the amount of feedback she can give you, because she’ll have to stop part-way through checkout. Instead, use solutions such as giving the tester a prepaid Visa card (at UserTesting.com, your account manager can help with this), or setting up a test server so you can fully test things like PayPal checkout without money actually changing hands. Of course, be conscious of the impact that using “fake” money has on the tester’s behavior.
• Use REAL customers when appropriate. If you’re testing anything related to personalization, account management, or purchase history, you’ll of course want to use existing customers. This is preferable over fake accounts you’ve created; you want the tester to see his own data. You can use intercepts for this, where users are recruited as they shop (if you don’t have a method for that, let one of our enterprise account managers set you up), or conduct a private panel test, where you invite your own users by email or social media.
• Use some REAL tasks – as defined by the tester. It’s tempting to write every task in detail, but there’s a danger in selecting all of your own tasks (see Task-Selection Bias: A Case for User-Defined Tasks by Richard Cordes). Don’t make the entire test artificial by telling the tester every step to take. Instead, let the user tell you what he wants to do, and watch him try to do it. You might discover an opportunity for improvement that you never would have thought of yourself.

Security of Personal Data

Return Policy and Guarantees

Customer Service Availability

Price / Value

Order Accuracy

• How confident are you in providing your financial information and contact information to this company?
• Is there a store/company that you trust more? Why do you trust them more?
• Was anything about the checkout process unclear or confusing?

The convenience of your checkout process can’t be verified by analytics alone. The customer might get from page 1 to page 3 quickly, but did he accomplish everything he wanted to, and did he hit any confusing points along the way? Qualitative testing is the next logical step after you have your quantitative data.

Here are several points along the checkout process to test for convenience:

More than 1 in 5 visits to your store are coming from smartphones and tablets, according to Monetate’s Q1 2013 eCommerce Quarterly report. And their shopping experiences haven’t been good, according to a recent survey. What’s more, 33% of them will immediately jump to a competitor’s site if they have problems on yours.

It’s definitely worth running some mobile user tests on your checkout process to make sure those customers don’t encounter any problems that make the checkout process difficult or impossible, such as problems with your forms, or elements such as modal dialogs that can be difficult or impossible to interact with on small screens.

You can get more testing ideas from the research paper The Four Mobile Traps: The Most Common Mistakes Made by Mobile Apps and Websites. The results and tips in that study came from observations of thousands of mobile usability tests.

Larger touch targets, larger text, and larger images make the checkout experience on Walmart.com feel much easier than the experience on ToysRUs.com.

The New Customer Experience

It’s no secret that offering a guest checkout option improves conversion (a 2008 Forrester report is bluntly titled “Required Registration Lowers Online Conversion Rates“), so you’re probably already postponing the optional account creation process to the end of the transaction like the top retailers do. But it’s still worth watching several new users trying to check out. Do they have any problem finding the guest checkout option? Are you making it as easy as possible for them to create an account with you by the end of the transaction, optimizing your chance to add to your registered customer database? Do testers see any compelling reasons to register?

So you’ve gone through the work of accepting 50 different methods of payment. Kudos. Everyone should be happy now, right? Well, maybe. Watch and listen to users trying to decide among the options. Have you violated the “simple trumps complete” rule?

Are your payment choices making it easier or more difficult for the customer to continue the transaction?

Remember the day when you first saw a “you might also like” section on a shopping cart? As a store owner, you thought you had died and gone to heaven, right? Now that feature is ubiquitous, although it’s not always used judiciously. When done right, upselling is not only a revenue generator for you, but a convenience factor and positive experience for your customers. Gather tester opinions to find out whether your upselling is contributing to their convenience, or whether it’s just getting in their way.

In this example of a business card purchase, the customer is presented with a large list of optional add-ons in step 2 of the checkout process, and a continue button is nowhere to be found above the fold.

If you have customers from other countries, have you streamlined their experience? Can they easily get prices in their own currency? Does your form validation allow them to enter their addresses and phone numbers, or do they encounter errors? You could try to account for all of the needs through a ton of research (which is still a good idea), but your testers might raise an issue you haven’t thought of. Use analytics to learn where most of your international business is coming from, and recruit testers from those countries (or use your own customers).

Errors are going to happen, and your error handling can either help the shopper get back on track toward completing the purchase, or cause them more confusion and frustration.

While user-defined tasks are helpful and were mentioned earlier, you might not want to wait for a user to encounter an error on their own, so come up with a list of errors you can force, and then lead the user into the error without telling them what’s coming. Watch their reaction and their recovery.

The previous page showed this laptop to be in stock, but after clicking Add to Cart, the shopping cart page shows a subtotal of $0, and there’s nothing in the cart. But the message explaining what happened is easy to miss if the buyer’s attention is drawn to the right side of the page where price and the continue button are normally found.

The gift giver has unique needs during shipping, such as the ability to add a personalized note, the option to have the gift wrapped, and perhaps the need to ship items from one order to multiple addresses.

Cabela’s does a great job of making gift options clear and consolidated (but you still can’t ship different items to different addresses).

The browser back button is one of the most commonly used navigation elements, yet some shopping carts fail to support it well. If a user clicks the back button, is any of his information lost? Do odd error messages or browser popups (about resubmitting data) appear that could confuse people?

Locking customers into the checkout process, and helping them focus on the task at hand by removing distractions, has its advantages for sure. But make sure such techniques aren’t interfering too much with convenience. Can a customer who decides to go back and shop for one more item find out how to do it? Are there any items you’ve removed from the interface that customers want back, and if so, does customer opinion and convenience outweigh the benefit of lock-in? Test customer behavior with different amounts and types of information on screen to see which mix works best.

Amazon aggressively locks buyers into the checkout process, providing virtually no way to get back to the store if a buyer decides to add an item.

• How could we have made this [faster / more convenient / easier] for you?
• Did anything frustrate you during the checkout process?
• If you had to make this purchase again, would you use this store, or a different one? If you’d use a different one, why?

The Anatomy of a Great Shopping Cart

If these testing tips were helpful, you might want to check out the graphic The Anatomy of a Great Shopping Cart, which shows several of the best practices well-executed.