The Better Interface — The Same Animal, but a Different Beast?
Many times, we look at a website and think well “I can build a better interface for that, or add features that better suits me as a user.”
I’ve always been interested in improving interfaces of existing products, which is what captivated me about the crypto/web3 premise of building front-end interfaces around existing on-chain backends. In the web2 world, Zillow comes to mind as an example of creating a well-recognized interface for viewing MLS real estate listings. The idea of using the same data as a product but creating a better one reminds me of the old Nike Kobe System commercial.
For those that remember the iconic commercial, Kobe Bryant asks Kanye:
But are you a different animal but the same beast?
As Kobe evolved over his career, becoming different animals, he maintained his signature Mamba Mentality, remaining the same beast.
What I want to build is the same animal, but a different beast.
To build from the same data but apply a different mentality in the product.
In line with my automotive interests, we at Project88 recently launched BATRANK, which is built on listings from popular car marketplace Bring a Trailer (known as BaT). My goals were as follows:
- How can we improve the discoverability of relevant listings for users?
- How can we easily provide data analysis like market trends?
- What unique features can we add to help users find new listings?
Before we jump into how BATRANK addresses these goals, we must first understand the relationship between BATRANK and BaT. BATRANK wasn’t developed to replace BaT, but instead supplement current and potential BaT users to find relevant listings for them. BaT is designed to buy and sell vehicles, which isn’t a proposition for BATRANK.
Goal 1: Finding the right car listing for you
When looking at BaT we can identify three core user personas:
- The buyer: People looking to buy a car
- The seller: People selling their car
- The spectator: People enjoying looking at interesting cars with no intention to buy or sell, usually commenting and sharing
While every user on BaT is a spectator, the reason why someone wants car listing data is different per persona. A buyer wants to know what cars are available and if they should buy it at a specific price. A seller wants to know how to maximize value for a successful listing. A spectator wants to find listings that are entertaining to them and participate in discussions with others.
The art of meshing these three personas in the same interface was achieved by improving the common denominator: discovery.
To improve discovery, we wanted to maximize the chances that a user can find a listing of interest as soon as possible. Let’s look at two of the discovery features we focused on:
First, we created a weighted popularity score as seen on the front page.
The premise is that popular listings will have a higher chance to be of interest for the average user if they were not searching for a particular listing at the beginning.
The second feature was the search. While search is available on BaT, we wanted to leverage our tech expertise to create a more robust experience. Using NLP, we were able to upgrade the search experience to allow users. The dropdown was also updated to give user more immediate feedback that their search query matches their intentions.
Goal 2: Everyone can be an analyst
To best provide a good user experience, we want to make sure the auction data is easily accessible and more importantly digestible.
For our MVP, we did not want to add data visualization as it requires user interpretation to gain insights. Instead, we thought about how to visibly present the data so the user felt like they gained a valuable insight immediately. To accomplish this, we used metric rankings for a listing on its listing page and badges like appreciating model and good value to identify trends.
Goal 3: I’m Feeling Lucky
With the plethora of car listing dat available, users wants a polished journey to take them from interesting listing to interesting listing. Whether this is from the home page or from a relevant listing, the journey needs to continue in a seamless manner.
One method we implemented was to address of a user scrolling down the popular listings and not clicking one. We decided not to use an infinite scroll, therefore we needed a mechanism to segue a user to a listing before they hit the bottom. So we added an I’m Feeling Lucky button that appears after scrolling to send the user to a random listing.
While this does not always lead to the most interesting listing, it is important that we explore different paths to help the user find satisfying listings.
Next Steps
While we answer the amount of feedback from our early users, our next steps are to provide deeper insights into the data for the user so they can create their own novel analysis.
There are other features like top comments that we added for our personal amusement as those are core to the BaT community and experience. If you have any feedback or interested more about BATRANK please check out batrank.com or send us an email at info@batrank.com. Happy car hunting!
About Project88
Project88 builds at the intersection of car culture, motorsports, and technology. From startups, tech giants, and the racetrack, we draw upon our experience to craft unique products and experiences for consumers and companies.
About the Author
When taking a rare breather from the world of crypto, you can find Kent searching for track days that fit his booked work schedule and trying to optimize the “it should work” aero on his track car.
