Folio (prev gudwork).
Optimizing onboarding flow and empowering first time freelancers to apply to projects.
Role
Timeline
Aug 2022 - March 2023
Team
Cathy Huang (Founder) // Sukhita Karthi (Lead Designer) // Kayln Kwan (Intern) // Rayna Shah (Intern)
Skills
| tldr
Context
Folio (previously Gudwork) is a freelance marketplace that connects college students with flexible freelance opportunities from startups.
I worked as a product designer with the aim of increasing the student user base and empowering first-time freelancers to apply for projects.
I led the user research, problem assessment, and final design recommendations, taking the lead designer’s components and design system into account. My role extended into a PM internship, where I collaborated with PM, Dev, and QA to approve and implement my proposed features.
| Results
Outcome
Implemented a modified onboarding flow in Q1 2023. Prioritized educational features on the homepage ahead of a summer 2023 marketing push, with 80% currently deployed.
Post-implementation results one year later include:
Growth of the freelancer community from 27k to 40k
35% increase in sign-up completion
1,300 vetted, active students onboarded.
| Solution
Educational resources + ease of onboarding = active freelancers
My team struggled to approach how to solve this problem, as we were trying to compress two separate reasons why freelancers were inactive into one solution. We identified two key factors for enabling freelancers to successfully apply for projects: 1) confidence in their skillset and application outcomes, and 2) ease of onboarding to the platform.
Our final solution(s) and the specific goals they target:
1) Streamlined onboarding flow, including questions to help pre-fill profiles and personalize resources (confidence)
2) Dynamic educational resources, including a chronological series of “How-to” articles and toolkits (ease of joining)
3) Community feed that displays achievements of novice freelancers and enables team projects (confidence)
| Problem
High drop-off rates during sign-up and inactive new freelancers
When I joined Folio, there was a discrepancy in the volume of freelancers and the volume of positions due to the founder’s close proximity to VC communities. My team was tasked with analyzing why this was the case and why the majority of first-time freelancers would create accounts but not apply for projects.
And so, we asked ourselves:
Initial research question
How might we empower and equip students with the right resources to freelance with ease?| Kickoff
How we approached the problem
Our main goal was to find out why and where freelancers were dropping off during onboarding and empower them to begin a freelancing project. As we had been freelancing on the platform for ~ 3 months, we decided to use our own experience to evaluate the current user onboarding flow and identify any bottlenecks that cause drop-off.
Initial flow (20 steps) ~ click to enlarge
We tried putting ourselves in the shoes of novice freelancers who lacked confidence or direction as to what they hoped to provide. I found it helpful to envision different user flows and predict when these hypothetical personas would drop off, depending on their goals for the platform.
One thing we noticed was that a highly skilled novice freelancer possessed the initial knowledge to start a project but not the freelancer-related know-how, such as talking to clients. This would cause a drop-off during the freelance-specific questions, like asking for hourly wages.
We also timed how long it took to finish signing up each time we went through the flow (omitting email, password, and 2FA steps). It took on average 5.97 minutes to complete, the bulk of which was spent adding experiences and reading through the 16+ selection choices for services & skills.
Why did we do this? To get a better understanding of pain points, confusion, and areas where improvements are needed. Onboarding is the first step to becoming a successful freelancer, so to enable a smooth user journey, the onboarding flow must also be streamlined and require minimal effort.
From this evaluation, we found that freelancers faced two sign-up hurdles:
Confidence Bottleneck: Uncertainty about freelancing services and pricing hindered progress (too many choices provided).
Effort Bottleneck: Overcoming this barrier demanded external aid or substantial time/effort investment.
After gaining an understanding of why and where users were dropping off, we began conducting primary and secondary research to validate (or invalidate) our hypothesis:
Students feel
choice overload
at key decision points*, thus preventing them from completing their objectives
*deciding services to provide, skillset to add, projects to apply to
| Methods
Interviews
We completed 36 user interviews, each lasting 30 minutes, with college students nationwide (freelancers & non-freelancers). Our outreach involved scripted questions and observed onboarding walkthroughs. We investigated why some users hadn't started projects and key obstacles they encountered in their journeys.
Affinity Maps
To consolidate findings, we created an affinity map to steer in a specific problem space. We focused on drawing out salient student habits and bucketing insights.
Students often stumble into freelancing via referrals, lacking formal guidance. This places them in an unfamiliar territory of delivering high-quality work for experienced adults, which can be daunting.
Thus, we found that (1) students seek external reinforcement before applying for projects, as a second opinion builds the confidence and knowledge many lack early in their careers, and (2) they express a need for a structured approach to navigating the freelancing ecosystem effectively.
| Synthesis
Forming Insights
In light of these discoveries, we categorized our research into four key insights, providing the foundation for launching our ideation process.
Structured Resource Needs
Students are seeking comprehensive, step-by-step freelancing guides tailored to their expertise and field, yet struggle to locate consolidated relevant resources.
Community & Support
Limited exposure to peers at a similar freelancing level leaves students susceptible to imposter syndrome. Establishing a community grounded in shared experiences will foster success.
Uncertainty & Confidence
Students often feel daunted when initiating their freelance journey, resulting in uncertainty and a lack of confidence in their skillset.
Effort Investment / Usability
Students are hesitant to dedicate effort to smaller tasks lacking clear objectives and immediate rewards. When faced with extensive information requirements, they abandon sign-up.
With these insights, we refined our How Might We statement:
HMW establish a structured support foundation that enables students to pursue projects with confidence?
After voting with the founders, we opted for an onboarding quiz system that uses user inputs to dynamically display personalized freelancing resources on their homepage. This choice engages users right from the sign-up stages, with the added advantage of requiring relatively minimal development efforts. In the long run, empowering users to become successful freelancers helps drive higher project completion rates and advance Folio's growth KPIs.
Following account sign-up, users will be prompted to take this quiz with an opt-out option.
We considered the following design requirements:
Maximum of 15 choices per question (reduce decision overload)
Links to external resources if asked to provide freelance-specific information (ex. hourly rate)
Skip option (user control)
Context dependant next steps
Preference questions (‘Skills interested in learning’) before information questions (‘Add education’)
With this feedback, we began brainstorming how to improve the flow of the onboarding quiz and certain requirements we gathered.
| Ideation
We initiated an ideation and voting process to brainstorm ways of establishing this robust support system.
| Explorations
Initial low-fidelity explorations
We sketched out our initial ideas to visualize the flow of the onboarding quiz and features we wanted to include on the homepage.
Iterations to the user flow
Apart from minor visual and hierarchy adjustments, user testing uncovered inefficiencies in the onboarding quiz flow:
Users appreciated the emphasis on skills rather than services but expressed concerns about students experiencing imposter syndrome when selecting skills they were ‘confident’ in.
They recommended reordering questions to prioritize preferences (like skills and help) over informational inputs (e.g., hourly rates) to make the process less daunting.
There were doubts about the quiz's ability to autofill profiles due to the limited information it gathered.
Mid-fidelity Concept Testing
With our initial idea conceptualized, we brought our design to six current freelancers and three students.
| Solution
Onboarding
Personalized to you
Complete now or later
Access tailored resources
Follow the pathway you’ve already started
The Student’s Freelancing Guide
| Reflections
Outcome/KPIs
Currently implemented (slightly revised as of 2023) into the platform
New onboarding increased sign-up completion rate by 35%
Newsletter / Community grew from 27k to 40k one year after implementation
1300 vetted, active students on the platform
Learnings
One of the most significant takeaways from this experience was the realization that designing within a product ecosystem necessitates a more comprehensive perspective. It's not just about designing a single, isolated feature; it's about understanding how each design decision contributes to the overall success of the platform. I had to consider the broader context, including business value, feasibility, and the impact on various user segments.
Additionally, drawing from my experience as a freelancer, I naturally gravitated towards ideation that addressed pain points occurring later in the typical freelancer journey. This experience taught me the importance of setting aside my personal biases and shifting my focus toward enhancing user confidence rather than solely promoting project completion.
Wishes
With more time, I would’ve loved to experiment with adding editable templates natively integrated into the platform, to save freelancers time building their portfolio and encourage them to send their first application to a project. This feature was deprioritized due to extensive development estimates, but I would be excited to see it in the roadmap in the future.
I also wish I had allocated more time for several additional rounds of user testing with the high-fidelity prototype, allowing us to closely observe user behavior on the platform and make informed enhancements to its overall usability.
