Portfolio
Prolancing (Solo Founder Project)Product DesignVisual & Branding

Prolancing: A 3-Month Experiment in Building a Freelance Marketplace

How I designed a full-fledged freelance marketplace (wireframes, UI, mobile app, landing page) in 3 months—then shut it down because I couldn’t solve the cold start problem.

3 months

Time to Build

20+

Screens Designed

2 (BTO, ATO)

Modes Designed

Case study visual
Built withProduct DesignWireframingMobile App DesignLanding Page DesignAI Product DesignStartupFreelance Marketplace

Context

The working context

Team

KV

FOUNDER, PRODUCT DESIGNER (SOLO)

Koushik Venkatesan

Overview

Late 2023 – Early 2024 · 3 Months (Gap Between PickYourTrail & Freightify) Prolancing (originally called ConnectX) was a freelance marketplace designed to solve problems that platforms like Upwork, Fiverr, and LinkedIn don’t address: - Team projects (most freelance platforms are 1:1, not team-based) - Real-time AI-powered task management (track tasks, suggest workflows, speed up shipping) - Portfolio community (Behance-style feed for freelancers to showcase work + get hired) - Tinder-style networking (swipe to connect with freelancers/clients by geo + industry) - Competitive pricing (5% commission vs. Upwork’s 20%, Fiverr’s 20%) I built: - Full wireframes (20+ screens for client + freelancer modes) - High-fidelity UI (using AI design tools for rapid iteration) - Mobile app prototype (Figma, fully interactive) - Landing page (pitch deck-ready marketing site) After 3 months, I shut it down. Why? - No clear distribution (couldn’t crack cold start problem: freelancers need clients, clients need freelancers) - No funding (bootstrapped, couldn’t afford to run paid ads or build a sales team) - Market validation issues (freelancers loved the features, but wouldn’t switch without clients already on the platform) This case study is about what I built, what I learned, and why I walked away.

My expertise

Product design, wireframing, mobile app design, landing page design, AI product design, market validation

Problem

What needed to change

How do you build a freelance marketplace that solves real pain points (team projects, AI task management, portfolio discovery)—and launch it without distribution or funding?

I had a 3-month gap between leaving PickYourTrail and joining Freightify. Instead of job hunting, I decided to build something.

The Hypothesis: Existing freelance platforms (Upwork, Fiverr, Freelancer) have major gaps: (1) No team project support — most platforms are 1:1, but real projects need teams. (2) No real-time task management — clients and freelancers use external tools like Slack, Jira, Asana. (3) No portfolio discovery — Behance is for inspiration, but you can’t hire directly. (4) High fees — Upwork takes 20%, Fiverr takes 20% + buyer fees. (5) Poor networking — LinkedIn is for full-time jobs, not freelance gigs.

The Solution: Prolancing (originally ConnectX) — an all-in-one freelance marketplace with team projects, AI-powered task management, a portfolio community (Behance-style feed), Tinder-style networking, integrated communication, and 5% commission vs. 20% on Upwork/Fiverr.

The Challenge: How do you design and build a product this complex in 3 months—with no funding, no distribution, and no existing user base? Spoiler: you can’t. But you can learn a hell of a lot trying.

Questions we had to answer

01

What is the “cold start problem” in marketplace businesses?

02

Why did freelancers love the features but wouldn’t switch platforms?

03

What would I do differently if I built this again?

04

How did AI design tools help me ship faster?

3 months

Time to build

Gap between PickYourTrail and Freightify

20+

Screens designed

Wireframes + high-fidelity UI for client + freelancer modes

2 (BTO + ATO)

Modes designed

Before the Order (job posting, applications) + After the Order (task management, deliverables)

$0

Funding raised

Bootstrapped, no VC funding

Cold start problem

Reason for shutdown

Freelancers need clients, clients need freelancers—classic marketplace dilemma

Alignment

What success had to look like

Success criteria

This project didn’t “succeed” in the traditional sense (I shut it down), but success was about learning how to design and validate a complex product in 3 months.

I defined success as: (1) Ship a full product design — wireframes → high-fidelity UI → mobile app → landing page. (2) Validate the features — talk to freelancers and clients, get real feedback. (3) Learn market validation — understand why distribution matters more than features. (4) Know when to walk away — don’t burn time/money on a product with no clear path to market.

I hit all of these. The product never launched, but I learned more in 3 months than I did in 2 years at PickYourTrail.

Design themes

Speed Over Perfection

Used AI design tools to iterate fast. Shipped wireframes → high-fidelity UI in weeks, not months.

Market Validation Trumps Features

Freelancers loved the features, but wouldn’t switch without clients already on the platform. Distribution > product.

Knowing When to Walk Away

3 months in, I realized I couldn’t solve the cold start problem without funding. Shutting down was the right call.

Process

01Research Existing Platforms (Upwork, Fiverr, Toptal, Behance)
02Define Pain Points (Team projects, task management, portfolio discovery, high fees)
03Wireframe 20+ Screens (Client mode, Freelancer mode, BTO, ATO)
04High-Fidelity UI Design (Using AI design tools for rapid iteration)
05Mobile App Prototype (Figma, fully interactive)
06Landing Page Design (Pitch deck-ready marketing site)
07Talk to Freelancers and Clients (Validate features, understand objections)
08Realize Distribution Is the Problem (Cold start problem)
09Shut Down After 3 Months (No funding, no distribution, no path to market)

Constraints

The tradeoffs we had to work inside

01

3-month timeline (gap between jobs)

Had to design, validate, and launch in 3 months—or abandon the project.

02

Bootstrapped (no funding, no paid ads, no sales team)

Couldn’t afford to pay for user acquisition. Relied on organic reach (which didn’t scale).

03

Cold start problem (marketplace dilemma)

Freelancers won’t join without clients. Clients won’t join without freelancers. Classic chicken-and-egg.

04

Solo founder (no co-founder, no technical co-founder)

Had to manage product design, strategy, and validation alone. No one to share decisions or the load.

Behaviors encouraged

  • Sign up as a freelancer (create profile, upload portfolio)
  • Sign up as a client (post jobs, host events, build teams)
  • Explore portfolios (Behance-style feed)
  • Swipe to network (Tinder-style matching)
  • Use AI task management (track tasks, get workflow suggestions)

Behaviors discouraged

  • Switch from Upwork/Fiverr without seeing an active user base first
  • Pay 5% commission when free platforms like LinkedIn exist
  • Join without clear ROI (freelancers won’t invest time unless they see jobs)

Process

How the work moved from messy to shipped

This was a 3-month crash course in product design, market validation, and knowing when to walk away.

Phase 1: Research and Pain Point Definition (Week 1–2) — Studied Upwork (high fees, no team projects), Fiverr (gig-based, not project-based), Toptal (elite only, expensive), Behance (can’t hire directly), LinkedIn (full-time jobs, not freelance). Key insight: every platform solves one problem well, but no one solves all of them in one place.

Phase 2: Wireframing (Week 3–4) — Designed 20+ screens across 2 modes. BTO (Before the Order): job posting, applications, portfolio browsing, networking. ATO (After the Order): project management, task tracking, deliverables, payments.

Phase 3: High-Fidelity UI Design (Week 5–8) — Used AI design tools (v0, Uizard, Galileo AI): input wireframes → AI generates high-fidelity UI → refine. Cut design time from 4 weeks to 2 weeks.

Phase 4: Mobile App Prototype (Week 9–10) — Built a fully interactive Figma prototype covering onboarding, profile creation, portfolio browsing, job applications, project dashboard, and AI task suggestions.

Phase 5: Landing Page (Week 11) — Designed a pitch deck-ready landing page to explain the product and attract beta users.

Phase 6: Validation and Reality Check (Week 12) — Talked to 50+ freelancers and 20+ clients. Feedback: freelancers loved the features but wouldn’t switch without active clients. Clients were skeptical (“Why trust a new platform when Upwork already works?”). The cold start problem: freelancers won’t join without clients, clients won’t join without freelancers, and couldn’t afford paid ads to kickstart both sides.

Phase 7: Decision to Shut Down (End of Month 3) — Distribution matters more than features. Marketplaces need capital to subsidize one side until the network effect kicks in. Shut it down and joined Freightify.

01

Wireframing — 20+ Screens, Client + Freelancer Modes

Designed 20+ low-fidelity wireframes across 2 modes to map user flows and validate structure before high-fidelity design.

BTO (Before the Order): Client — post jobs, host events, build teams, browse freelancers. Freelancer — apply to jobs, participate in events, upload portfolios, swipe to network.

ATO (After the Order): Client — manage tasks, track progress, communicate with freelancers, review deliverables. Freelancer — view tasks, submit deliverables, communicate with clients.

Key screens: Home (Behance-style portfolio feed), Explore (job + event listings, freelancer discovery), Post/Build, Messages (Tinder swipe + traditional messaging), Profile, Projects (AI-powered task dashboard).

Wireframing — 20+ Screens, Client + Freelancer Modes media 1

Wireframe

02

High-Fidelity UI Design (AI-Assisted)

Used AI design tools (v0, Uizard, Galileo AI) to rapidly iterate from wireframes to high-fidelity UI: input wireframes → AI generates high-fidelity UI → refine colors, typography, spacing, components → build design system → export to Figma.

Cut design time from 4 weeks to 2 weeks.

Key design decisions: dark mode default (professional, reduces eye strain), card-based UI (Instagram-inspired, easy to scan), AI task suggestions (ChatGPT-style interface), Tinder-style swipe (networking feels fun, not transactional).

High-Fidelity UI Design (AI-Assisted) media 1

High Fidelity UI Mockup

03

Mobile App Prototype (Fully Interactive Figma)

Built a fully interactive Figma prototype covering both freelancer and client flows.

Freelancer flow: onboarding → browse jobs → upload portfolio → Tinder-style networking → manage projects (AI task suggestions, chat, submit deliverables).

Client flow: onboarding → post jobs → build teams (select freelancers from matches, send invites) → manage projects (assign tasks, track progress, review deliverables).

Key features: AI task management (chat-style interface, suggests workflows, SOPs, deadlines), Tinder-style networking (swipe right = interested, left = pass, up = view profile), real-time notifications.

Mobile App Prototype (Fully Interactive Figma) media 1

Mobile App Screenshot

Figma assets may take 1–2 minutes to load from CDN. Please wait if you see a blank or loading frame.

Figma Prototype — Mobile App

04

Landing Page Design (Pitch Deck-Ready)

Designed a landing page to explain the product and attract beta users.

Key sections: Hero (“Empower Your Freelance Journey with Prolancing” + CTA) → Problem (fragmented platforms, high fees, no team projects) → Solution (all-in-one marketplace with AI, portfolio community, low fees) → Features → Pricing (Free 5%, Prolancer $999/month no commission, Enterprise custom) → CTA (“Join Beta” email capture).

Figma assets may take 1–2 minutes to load from CDN. Please wait if you see a blank or loading frame.

Figma Prototype — Landing Page

Takeaways

What changed once the work met reality

What didn't go as planned

01

The cold start problem (freelancers need clients, clients need freelancers)

Classic marketplace dilemma: freelancers won’t join without active jobs, clients won’t join without active freelancers. Reached out to 50+ freelancers (most said “I’ll join once there are clients”) and 20+ clients (“I’ll post jobs once there are freelancers”). Couldn’t afford to subsidize either side. Lesson: distribution > product. Marketplaces need capital to solve the cold start problem.

02

Competing with Upwork/Fiverr (established platforms with 10M+ users)

Even with better features (team projects, AI task management, lower fees), freelancers and clients wouldn’t switch. Switching costs are high (profiles, reviews, existing clients on Upwork). Trust is earned, not designed. Network effects: Upwork has 10M+ freelancers—clients go where the talent is. Lesson: features don’t matter if distribution is weak.

03

Solo founder with no co-founder

Had to manage product design, market validation, and strategy completely alone. Every decision was on me with no one to challenge or validate it. Lesson: marketplaces especially need a co-founder with sales/BD experience to crack distribution. I could build the product but couldn’t crack distribution alone.

What shipped

Designed a Full Product in 3 Months (Wireframes → UI → Mobile App → Landing Page)

Shipped a complete product design—wireframes, high-fidelity UI, mobile app prototype, and landing page—in 3 months as a solo founder.

20+ (wireframes + high-fidelity UI)

Screens designed

2 (BTO: Before the Order, ATO: After the Order)

Modes designed

3 months

Time to build

Used AI Design Tools to Ship 2x Faster

Used v0, Uizard, and Galileo AI to rapidly iterate from wireframes to high-fidelity UI—cutting design time from 4 weeks to 2 weeks.

From 4 weeks to 2 weeks (50% faster)

Design time reduced

v0, Uizard, Galileo AI

AI tools used

Validated the Market and Decided to Shut Down

Talked to 50+ freelancers and 20+ clients, learned that distribution matters more than features, and made the decision to shut down after 3 months rather than burn more time on an unsolvable problem.

50+

Freelancers interviewed

20+

Clients interviewed

Distribution > product. Marketplaces need capital to solve the cold start problem.

Key learning

Outcome

What shipped and what it proved

3 months

Time to Build

Gap between PickYourTrail and Freightify

20+

Screens Designed

Wireframes + high-fidelity UI for client + freelancer modes

2 (BTO, ATO)

Modes Designed

Before the Order + After the Order

50+

Freelancers Interviewed

Market validation and feedback

20+

Clients Interviewed

Market validation and feedback

$0

Funding Raised

Bootstrapped, no VC funding

Distribution > product

Key Learning

Features don’t matter if you can’t solve the cold start problem. Marketplaces need capital.

What came next

This project didn’t “succeed” in the traditional sense (I shut it down), but I learned more in 3 months than I did in 2 years at PickYourTrail. Key learnings: 1. Distribution > product. A worse product with better distribution wins. 2. Marketplaces need capital to subsidize one side until the network effect kicks in. 3. Solo founders burn out faster. No one to validate decisions or share the load. 4. Knowing when to walk away is a skill. 3 months in, I realized I couldn’t solve the cold start problem without funding—so I shut it down and joined Freightify. What I’d do differently: - Start with one side of the marketplace (freelancers only, or clients only—not both at once) - Focus on a niche (design freelancers only, not all freelancers) - Find a co-founder with sales/BD experience I’d love to apply these learnings to a company with existing distribution and a clear path to market.

View more work
2026 — Kelashik.