Secondbuy is a project that aims to improve the second-hand bike retail process for both sellers and buyers on a college campus.
User research, Ideation, Wireframing, Interaction design, Usability testing, Iteration
Danrui Sun, Lois Zhao,
Gaby Shangguan, Akhil Mohanan
Research shows that it’s common for students to buy used bikes...
Buying or selling a used bike is always a smart choice for students because it allows them to save money and get around campus quickly, given their busy schedules. Our research shows that students often buy used bikes on Facebook Marketplace- in fact, it’s ranked as the third most common purchase category on the site!
...However, there are challenges with the current used bike sales process
Due to the fact that Facebook Marketplace was designed for a wide range of products, there are many design considerations specific to bikes that are not accounted for. For instance, on the buyer experience side, buyers lack easy ways to determine level of bike quality, as well as key features which often are not conveyed just by a picture. Due to this confusion, sellers become inundated with vast quantities of messages and feel overwhelmed when finding the right buyer.
Our group aimed to address these problems and help students have an enjoyable retail experience. To do so, we followed the user-centered design process to design an application called SecondBuy, which improves the experience of selling and buying used bikes.
Take a quick look at our design solution
Talking to users to understand user needs and context of use
To understand our user needs, we conducted deep research into users’ second-hand bike retail experience. We first created a survey and shared it at our school's Facebook group and second-hand bike Reddit group to get a foundational understanding of user needs. Next, I led observations and interviews, of both buyers and sellers, at a second-hand bike store.
Focusing on understanding users' characteristics on a broad range
Focusing on behaviors and feelings of users during bike selecting process
Focusing on synthesizing user's quotes into common themes
Contextual inquiry at second-hand bike shop
Research synthesis session
To better analyze our survey and interview result, we conducted a research synthesis session to discover the most common user problems. We also created an affinity mapping diagram to identify common themes of users’ behaviors, thoughts and needs.
Communicating our two types of users: the seller and the buyer
After synthesizing our research findings, we identified the goals and pain points of our target users. Based on these findings, I created the personas of both buyers and sellers. The two user personas of our target groups also helped us focus on specific findings to address with our design solutions.
Key finding 01
Sellers lack the bike expertise needed to price their bike correctly
Through talking to bike experts, we found that the prices that sellers were listing their bikes on Facebook Marketplace at, were significantly higher than the bikes’ true market value. This price mis-match often led to a long back-and-forth negotiation between buyers and sellers. If sellers had a better understanding of their bikes’ true market value (given condition, key features, size, etc.), then this wasted time and effort negotiating price could have been avoided.
Key finding 02
No standard in place for capturing all of the bike information that buyers care about
Buyers often grew frustrated because the information they cared about, such as bike size, weight, features, an quality, were not clearly captured in the posts. Therefore, they often spent a lot of time and effort inspecting the bikes themselves, prior to making a purchase. Or worse, some buyers purchased a used bike from the site and then later regretted it, upon realizing that the bike was different than expected.
It’s time consuming for buyers to try out and compare bikes
Even in cases where the sale posts accurately depicted bike quality and features, buyers still spent a lot of time trying out bikes for test rides from different sellers to find the best bike for them. In many cases, buyers grew exhausted from the travel and time required to do this comparison.
Key finding 03
Sellers also spent a lot of time and effort to sell their used bikes
For example, sellers had to schedule showings, negotiate the bike price with potential buyers, and reply to buyer questions about their bike. After all of this work, there still remained possibility that the sale would fall through, in which case, the seller would have to repeat this process！
Converting research findings into design directions
Following our research, we met as a group to imagine solutions to address the user needs that we identified.
Creating solutions based on user's need
We further organized brainstorming sessions for the three user needs. We generated over 30 different design ideas across our 3 user needs. I sorted each idea by what problem it solved and created the idea map, shown below:
Choosing our design ideas
In order to prioritize our 30 design concepts to select the top 3 ideas, we considered 1) functionality, 2) feasibility, and 3) effectiveness in solving for user needs.
Setting up an expert community is effective in addressing all three user needs. Think 'Uber for experts' where people who are knowledgable about bikes are asked to evaluate the bike that the seller is trying to sell. Then experts are paid by commission of each bike they examined. The more bike they work on, the more money they will earn.
Video-based product listing
The bike product page will show a 360 video of the bike and the test video of the same expert riding different bikes. Users can see close up photos of the bike to get a sense of the bike’s condition and size.
Users can use the size calculator to see if the bike is a fit. Upon input of their height, a figure of the input size is shown to scale next to the bike. So that users can figure out if the bike would be suitable or not.
Considering user journey when implementing the design concepts
After thinking deeper about the design ideas, We realized some shortcomings. Such as: it would be troublesome for experts to go to different places to check and deliver the bike. Then, I reflected on our target users’ needs. Both of them are on-campus students and will come to school every week. The school is an ideal transit hub to exchange the bike. So, I proposed the concept of an on-campus bike transit station to improve our design idea.
On-campus bike transit station
Secondbuy will install bike racks with smart locks at multiple locations around campus. Sellers can drop off their bike with QR code at a specific place for their convenience. Buyers can pick up their purchased bike with QR code at the transit station anytime.
Below is the overall concept model:
After thinking deeper about this solution. I realized that we are not just trying to solve users’ problems. We are actually providing a novel service that could generate money. Assuming we are building a startup called Secondbuy, we furthered discussed a business model to support it.
Creating user flow
Based on the design ideas, I synthesized one design solution to make it work as a whole. I created this user flow to show the interaction between different stakeholders.
Creating pixel-perfect design solution
Buying a bike
On landing the page, buyers can see products that are on sale, or those with recently reduced prices. After arriving on the product detail page, they could view: the bike video, expert reviews, and bike metrics. Additionally, they can use the size calculator to find an ideal bike. When they check out, they could use the QR code to get the bike at on-campus transit station.
Selling a bike
Since the buyer could be the potential seller, our design integrated the selling and buying feature in one solution. We tried to have the selling process as easy and quick as possible. The sellers only need to enter basic bike information and drop off the bike at on-campus station. The rest of information will be completed by the expert.
What if some edge cases happen?
Design is not just thinking about solutions in an ideal scenario. In real life, some edge cases will occur. After creating the main user flow, I kept thinking about how to push the boundary of our design. For example, what if the buyer is not satisfied after unlocking the bike and trying it on? The solution should also enable buyers to return the bike and get refund. Buyers should be given enough time (typically 7 days) to try the bike and ensure it’s in good condition, and return it if needed.
Getting feedback from users
To test the usability and general concepts of our design, we conducted 8 usability testing sessions with our target user group. We used the think-aloud protocol to understand whether users can use our design solution as intended, and also get qualitative feedback on their impressions of the experience. Conducting usability testing helped us confirm our design choices and find opportunities for continued improvement.
Iterate design based on feedback
Issue 1: Users are unclear about expert role and examination standard. Some of the comments we heard frequently from participants were that they did not know what was the role of the review and they are concerned if the listed tags are all the things examined. We also decided to highlight key information from the expert checklist more clearly.
Based on users' need. I designed a new version to highlight expert's standardized examination of different parts of the bike. However, this design took nearly double the place of original design. I then designed another version to make a balance. As users care more about problems in the bike rather than the good parts. Therefore, I emphasized the guarantee of quality of bike sold at Secondbuy and only list the concern parts.
Issue 2: Participants did not understand the function of the edit button on the wishlist. After talking with participants, we discovered that they didn't expect too much function to modify their wishlist. They just want the basic function of adding and deleting items in wishlist.
I first changed the edit button into delete button to make it intuitive for user. However, after thinking deepr about this design, I found it would take 3 clicks for user to delete an item: click delete button -> select delete item -> confirm. Then I further designed a new version which only took 2 clicks to finish the task.
Key learnings and my thought after project
Combine business model with design: During our design process, we found to make the design reasonable we need to consider the business model first. For example, we need to consider how to pay the expert and how should the return policy work. Bad business model will also harm user experience. Motivated by this project, I took a business class: Analysis of emerging technology to gain better understanding of the market and more business sense.
Keep thinking and iterating: After finishing our main user flow. I kept reflecting if the design meets our target users' need. If there is anything missing? Then I realized that there were some edge cases I missed. I kept iterating on the design to handle the edge cases. However, in usability testing, users still posed some edge cases we had never thought about. This experience motivated me continuously thinking and iterating my design.