#ModevUX: Multiscreen Design @ Ebay

Not just about different screens, focusing on holistic experience of getting inspired about a product, transacting, receiving. hopefully a good enough experience that you’ll come back again again.

If you focus on moments and not how they are all connected, things are going to break down pretty quickly.

40% of ebay transactions go across multiple screens with multiple touch points.

What do we need to be doing when we’re designing these products?

Model: Born out of Research

Morning: phone, tablet. Daytime: computer, phone. Evening: computer, phone, tablet; peaks after dinner. Important when you think about shopping, this is when most transactions occur. When users are engaged in product research, they chip away during the course of the day, and have more intent at night.

What is the role of User Research?

To the end of the user research cycle, take a tactical view. Do a lot of research at ebay, so this is very important. However continually working much further up the stream – seed the org, product teams, with insights, so we are moving forward and not just keeping things in the status quo, not just getting one preconceived product as good as it can be.

What are users trying to achieve agnostic of device? What is their context?

Consumer to consumer: less engagement with sellers on the mobile size. Dropping off, not what expected. Director on product wanted to figure out what was wrong with the UI. Hadn’t taken the time to understand how mobile was playing a part in consumer to consumer testing. So started with field research.

There’s a very established PC listing ritual. It’s a tedious process, even harder on smartphone. Conducting product research was considered a challenge on the small screen. Creating a listing involves multiple windows, searching multiple products. Lots of difficulty of typing on a small keyboard. Editing is even more tedious on a smart phone. It’s important to creat fluid transitions between devices. Not just about UI, but about the multi screen ecosystem.

RITE studies in US and DE. People starting a listing, take a picture, save it, then go to another device and continue it from tablet/PC.

Have to step back, think of this agnostic of device. Without that, wouldn’t have got there, or at least not as quickly.

Searching on smart phone enables immediacy. If you can’t find a product, can’t buy it. Ebay is at a disadvantage here compared to other companies, have a large database, but it’s messy. Not like amazon, not a catalog. Listings created by merchants, and consumers. They don’t put in all the details that an amazon listing has. Search is really important for finding, and refining to get to what you want.

How do we enable immediacy? Because that is what you want on Mobile.

Triangulate the data that you do have, communicate to your product teams. The consideration – what’s different about the mobile use case? You’re in a distracted environment, you’re moving quickly. How does that fundamentally change the way that we need to think about search on mobile?

Make sure all people on the team have a good understanding, then do some ideation. Came out with tons of different prototypes. Built on html5, using angular. Different swipes to rethink search. Then tested using “compelled tasks” – the user’s own tasks.

When you are really looking for something there are whole other behaviours that are really important to see.

Show most used refinements directly on the screen in a carousel.

If hadn’t started early, taken the time to ideate and understand what already know wouldn’t have built the right thing, as it was, took at least 6 months.

See a huge increase in usage of refinement. Still doing testing, but usage of refinement is way up, and that was the goal.

User Research Takeaways

• Start early and go wide.
• Seek to understand user behaviour and perceptions agnostic of device.
• Triangulate existing data and insights.
• Test using natural “compelled” tasks.