Journi is a rating system that encompasses the entire client-freelancer work journey on freelancing platforms.
Tasks
User Research
UX Design
Timeline
September - October 2021
(4 weeks)
Team
Eileen Yan (me!)
Jak Sirimongkolkasem
Nicole Li
Context
Class project for CGT 522: UXD Fundamentals at Purdue University
Background
With the rising “gig economy” for designers and developers, freelance work has become a streamlined process. In freelancing platforms such as 99designs, Fiverr, and Upwork, freelancers are evaluated by their past clients, most commonly through the 5 star rating system. However, this rating system does not always capture the overall client-freelancer journey working together. My team’s goal was to design a rating system that alters the way freelancers are evaluated, so that they receive a fair review of their work based on their entire experience. Freelancers will then be able to use this system to improve upon their professional skills.
My Contributions
Secondary Research (Comparative Analysis)
2 User Interviews
Affinity Diagramming
Mental Model Diagramming
Personas
Scenarios
How Might We's
Problem Framing
Wireframing & Prototyping
3 Usability Tests
Deliverables
We designed low fidelity screens of Journi, a stage-based, emoji-style rating system that would be implemented into the freelancer-client chatbox on a freelancing website.
EXPLORATION
How are Current Freelancers Rated?
Top freelancing platforms use a 5 star rating system to evaluate freelancers. However, most of these ratings are optional, which means that clients can easily forgo their ratings altogether. Personas and mental model diagrams helped us determine the perceptions and experiences that freelancers and clients had when using freelancing platforms. These were based on secondary research and six user interviews.
Meet the freelancer, Christine
Christine prioritizes improvement and needs a rating system that allows her to grow via constructive criticism. Her current pain points are that:
she does not have enough reviews.
one negative review will impact her entire reputation on the platform.
she does not understand what the client ratings are based on.
Meet the client, Andy
Andy would like a rating system that helps him best filter out the many available freelancers while being quick and easy to complete after receiving the work. Furthermore, he:
does not completely trust the 5 star rating system, which prompts him to go into freelancer profiles to read through previous client reviews.
still has a hard time differentiating between quality designers despite the rating system.
PROBLEM FRAMING
Freelancers Can’t Pinpoint Areas for Improvement
Clients typically give short and vague feedback for good to average work, which is not always beneficial to the freelancer. Because of this issue, freelancers are unable to properly improve upon their work, which results in repeated mistakes. Since many freelancers also prioritize growth, this setback can be detrimental to their overall career progress. We made sure that our design reflects the freelancer - client project journey, so that freelancers can determine specific areas of improvement.
Freelancers Fear Negative Reviews
Freelancers are afraid of receiving negative reviews that could impact their overall score, and thus their reputation on the freelancing platform. In some instances, this pushes freelancers to provide additional work to guarantee a more favorable review from the client. To account for this issue, we focused on a design that would allow freelancers to receive constructive feedback but not be penalized by negative reviews.
IDEATION
Initial Wireframes
After several rounds of brainstorming, we evaluated our ideas based on previously determined design criteria and our problem frame. We ultimately decided to move forward with a stage-based rating system that would be implemented in the chatbox between freelancers and clients. Initial representations of these ideas can be seen in the wireframes below.
To address our first design goal: the rating system should reflect the freelancer-client’s journey of working together, we proposed using a staged rating system, where the client rates the freelancer’s performance and their experience at each stage.
To address the second design goal of providing meaningful feedback for freelancers to improve, we proposed:
attaching meanings to each score by providing a definition for each rating.
using emojis instead of number/star scales to convey sentiments and emotions of the client’s experience working with the freelancer.
To address the third design goal of encouraging critical and targeted feedback while not harming freelancer’s future opportunities, we proposed a "private note" feature. This note will give clients and freelancers a chance to communicate any misalignment or miscommunication.
To ensure the altered rating system employs a quick and easy mechanism for clients, we proposed using a sentence builder for leaving written feedback.
Sketches of the Client’s Journey
The client’s journey of browsing the freelancing platform and trying to find a freelancer to work with.
What the client will be able to see and do once they’ve selected a freelancer’s information page.
The process of how the client can give feedback to the freelancers.
TESTING & ITERATION
We conducted two rounds of testing, including concept, guerilla, and usability tests, with a total of three freelancers and four clients. Our insights and final prototype can be seen below.
Staged Review Process
Within the freelancer-client chat on a freelancing website, a staged review process helps freelancers pinpoint where they can make improvements in their work.
Clients are prompted to leave feedback after each stage has been completed. This way, freelancers receive constructive feedback to ensure that they are meeting the client’s needs.
To account for the diversity of projects, the staged review system provides the flexibility for freelancers and clients to mutually decide when each stage starts and ends.
Once a stage is complete, the freelancer has the option to mark it as done, then the system prompts the client to leave feedback about their experience working with the freelancer.
Private Feedback
Freelancers found private feedback features helpful for them to view critical feedback without worrying about the impact that negative feedback might have on their reputation.
Freelancers expressed that they would like to have clients to leave feedback as much as possible, not just for misunderstanding or misalignment. Thus, our iterated design prompts private, optional notes after every stage of the project.
Yes/No Recommendation Rating System
At the end of the project, clients see an overview of their staged reviews. The overview of staged rating feedback is intended to help clients make the judgement of whether or not to recommend this freelancer.
Clients have the option to recommend freelancers to future clients via a yes/no recommendation button and write public feedback. The limited options avoids the tendency of extreme ratings that can arise with a five-star system.
To deemphasize the association between freelancers’ overall performance with a five star rating scale, we decided to show their recommendation rate ( # of clients recommended / # of total clients). We intentionally avoided providing the final percentage of recommendations so that experienced freelancers with many past clients would not have a lower recommendation score than new freelancers with just a few past clients.
FINAL PROTOTYPE
Our final prototype was created with a freelancer flow and a client flow.
Interact with the freelancer task flow
Interact with the client task flow
REFLECTION
Because freelancers and clients have different goals when using freelancing platforms, I had difficulty designing for both. Nonetheless, this forced me to carefully evaluate my design decisions to ensure that they solved problems relevant for both user groups. In terms of my team’s solution, while the use emojis helps users translate more subjective feelings into defined categories, it might be interpreted differently by people depending on background and life experiences.
With this design, my team made the assumption that the clients are incentivized to rate freelancers because rating them throughout the journey can directly improve their experience with the freelancers they have hired. However, we were not able to test this assumption due to the timeframe of the project. This would need to be tested in a real-world context, where clients are actually interacting with the freelancers for a paid product and influenced by external factors such as time and effort.