Company: Neighbor Storage Discovery & design: 2 months Team: Myself and Senior Leadership (Neighbor had no product managers at this time) Platforms: Desktop web, mobile web, iOS, Android
Neighbor is like Airbnb, but for storage and parking.
I analyzed 1000 transactions one-by-one to learn what renters usually store. That way, we’d know what insurance to offer depending on the items.
Me after looking through 1000 reservations
Additional learnings
If a boat or other item is on a trailer, we need to know about the trailer too
Users sometimes say "Car" when it's actually an SUV or truck
A lot of people store old vehicles. Some hosts complain about this. Should we require a picture of the vehicle if it's old enough?
Categories are subjective. Users may not associate some items with most common categories (items such as food cart, golf clubs, kayaks, tools, kites, snowboard, other hobby equipment)
Some vehicles don't fit in any category, such as fishing boats, converted busses, kayaks, food carts.
Design progression
Explanation
(Slides 1 & 2) Sketches and wireframes
(Slides 3 & 4) Early designs
(Slide 5) We had a lot of confusion about what information we needed to capture for liability reasons. This was an example of that, which we ultimately didn't build.
(Slide 6) Final designs
Here are a few ways that we incorporated research
We leaned harder into requiring renters to write out each item that they are storing.
We added a section to capture vehicle-specific details for insurance reasons.
We added a section that explains what our plans do or do not cover.
We created categories to identify what renters were storing.
User testing
Look at eBay, and you'll find endless categories or items. We faced the a similar problem in helping users to tell us what they are storing. This test helped us to create categories that people would understand without confusion.
I tested only a small portion of my designs on User Brain. I should have tested a lot more. In the end, the interface performed extremely well, but we missed out on this learning opportunity.
Final designs
Checkout flow
This project included a full re-design of Neighbor's checkout flow:
Result
Adoption exceeded our goal
We saw no drop in conversion.
A marketplace only makes a small percentage of each transaction, so this meant huge revenue for Neighbor!
I'm under NDA so can't share exact numbers.
Reflection
The user testing wasn't that helpful. Looking back, I really should have done more.
We were trying to do too much. It would have been better if I had not re-designed Neighbor's entire checkout flow. This was before we had product managers, so it was a lesson learned the hard way.
I spent way too much time making those vehicle graphics. I should have kept them simple and moved on.
I regret not doing more discovery user research in general. I did some, but we relied too much on internal opinions to make decisions. The design performed well from a business perspective, but I wish I had used my time better to know whether customers had a great experience.
