Usability Fail: The Seamless Redesign and User Autonomy

Here in Boston, my go-to online food ordering service is Foodler, but in New York City, it’s all about Seamless. These sites, and their competitors, allow you to browse local restaurants delivering to your address and place and pay for orders online. It’s a convenient way to order delivery food without needing to use the phone or worry about having cash on hand.

At the end of June, Seamless redesigned their site. Updates were needed as a result of the merging of Seamless and GrubHub (and folks who work in internet technology know the pain of needing to marry two different systems within a shared front end). According to the team at Seamless, the new design was selected as a result of user research and testing that suggests its superiority to the original design. CTO Brian Lanier said, “We wouldn’t have made a decision to move forward if it wasn’t in the best interest of the diners and the business.”

Except users hate it. Here’s a sampling of the headlines about the new Seamless:

• The Seamless redesign isn’t being well-received, on Eater
• What the hell is going on with Seamless? on Gawker
• Why Seamless’ redesign is turning people’s stomachs, on Wired

Regardless of the merits or demerits of the particular design, I think one reason why the Seamless redesign is so poorly received is that it fails to support user autonomy (which is an antecedent of motivation) in critical ways. Specifically, it:

• Changes the value proposition mid-relationship. Customers used to be able to do some things on Seamless (such as review and recreate previous orders) that are not possible right now in the new design. Changing the rules of a product’s use like this can be very demotivating for users.
• Deprives users of choice in how they use the product. The loss of the archived orders is a great example. Many people find a particular order they like, and will simply recreate it time and again with a few clicks. Now, they must start from scratch every time. They can accomplish the same objective, but are forced to use only one method of getting there.
• Fails to clearly map the user’s path. Not only has the site functionality changed, but users aren’t sure how to navigate the new site, and there isn’t any sort of clear guide to help them. Users have tweeted questions about basic functionality because they can’t figure out the answers themselves. Murky mapping is a poor autonomy experience.

Given my career working with web technology, I sympathize with the challenges Seamless no doubt faced in integrating their site with GrubHub’s. This couldn’t have been an easy task. And I believe that their usability studies did show that the new site design worked, at least for the tasks they asked test users to complete. But what seems to be missing is an understanding of how users interact with Seamless “in the wild.” The redesign fails to enable that type of interaction, and in doing so, also fails to support user autonomy. I’ll be eager to see what future releases bring for Seamless and if it can restore its standing in the hearts of users.