Melon
Kevin Wang was a frugal Georgia Tech student frustrated with the cost, unreliability, and complexity of food delivery. While working as a software engineer and pursuing a Master's in Computer Science, he noticed existing delivery companies were combining orders to reduce trips—but only 1-2 drop-offs at a time. He wondered: what if you could efficiently drop off 20+ items together? The challenge appealed to him because it required business acumen and customer validation, not just technical complexity. Atlanta's entrepreneurial culture had exposed him to many founders and approaches, making the timing right to test a new hypothesis.
Kevin and co-founders Jack Olinde and his brother Jeff started with the simplest possible MVP: a group chat. They broadcast menus of 2-3 items from popular restaurants with specific drop-off times and locations, targeting dense student populations on campus. They spammed group chats and cold-added friends to acquire users. Within one week, they validated the core insight: their 17-item delivery completed in 45 minutes proved that pooling worked and consumers valued cheaper prices despite limited selection and fixed times. They even made small margins using catering prices. Then COVID hit, sending students home.
Undeterred, they pivoted to apartments in June 2020, launching a website with automated daily text menus. Growth was slow—75 subscribers by summer—as their original customer base dispersed. They spent the slow months learning logistics, studying customer behavior, and building an online ordering and tracking system with card payments. When students returned in August, they relaunched with a polished product and saw rapid growth.
The first customers came through direct outreach to campus group chats and friends at Georgia Tech. The MVP's extreme simplicity—a group chat—meant zero friction for early adopters who already faced the problem. When they relaunched post-COVID, they used Reddit, group chats, and tested Facebook and Google ads. However, the channel that delivered the most consistent growth was organic: posting friendly versions of their menu to social media. Friends ordering together improved both user retention (shared experience during pandemic isolation) and trip efficiency (nearby orders clustered better). Restaurant diversity and quality became competitive advantages that encouraged word-of-mouth.
What worked: the core logistics model. By August-October 2020, Melon achieved $10K MRR with 500 users, averaging 15+ items per trip in under 30 minutes. Their best drop-off delivered 47 items in one trip. The fixed time windows eliminated the unpredictability of on-demand, allowing restaurants and drivers to prepare better. Customers saved thousands in delivery fees.
What didn't: unit economics at scale. Kevin initially estimated 15% margins over $150K MRR after paying restaurants and drivers. But actual data revealed a harsh truth: margins would approach zero even in the best case. Driver wages were higher than expected because the work—rapid stops, food management, constant communication—was harder than 1-2 order delivery. As orders grew, consistency and reliability declined, eroding the efficiency gains that made the model special. The path to profitability looked identical to Uber Eats and Postmates: subsidizing orders and burning cash to gain market share.
Kevin also noted they tried to do too much—limited drop-off locations, experiments with gamification, plans to add groceries and crafts. This scattered focus meant they learned about many concepts superficially but lacked deep knowledge about any single one. They didn't build deep relationships with restaurants; Kevin felt like a salesman rather than a partner.
In October 2020, after preparing to raise $275K in pre-seed capital, Kevin made the difficult decision to shut down. He believed the right choice was to decline to become "just another" food delivery startup. The burn rate to expand across cities would be unsustainable, and without a fundamental business model shift, they couldn't outcompete larger platforms.
Looking back, Kevin identified lack of startup experience as a critical weakness. He and his team were all technical and intelligent, but they spent too much time learning frameworks and not enough time deeply understanding customer segments, cuisines, and restaurant needs. His reflections offer candid lessons: start with an even smaller niche, seek out failure rather than comfort, build intimate restaurant relationships, and ensure team expertise complements the problem space. As of late 2020, Kevin was pivoting to software engineering, data science, and machine learning roles.
- •Fast MVP validation in a low-technical-risk domain allowed them to test core assumptions (demand for pooling, customer willingness to pre-order) within one week, giving them confidence to persist through COVID pivots.
- •They solved a multi-sided marketplace problem elegantly by aligning incentives: pooling reduced driver miles and time, restaurants got volume commitments, and customers paid less—making word-of-mouth and social sharing organic growth drivers.
- •Focusing on a hyper-local, dense customer base (campus, then apartments) during initial growth allowed them to achieve the efficiency metrics that made the model mathematically work at small scale.
- •Pandemic isolation created unexpected demand for shared experiences (groups of friends ordering meals together), which they leveraged to both improve logistics (clustered orders) and user retention.
- •They failed because they did not validate the path to profitability at scale before committing to fundraising; the unit economics didn't improve as volume increased, forcing a hard choice between becoming a VC-dependent company or shutting down.
- 1.Start with the simplest possible MVP (a group chat works)—test core assumptions about demand and willingness to adopt before building software, which is a sunk cost.
- 2.Target a small, dense geographic niche where your unit economics actually improve with volume (not one where you need subsidies). Validate margins before scaling; model out driver costs, restaurant commission rates, and customer acquisition costs in Excel before raising capital.
- 3.Build one-to-one relationships with restaurants as true partners, not product vendors. Understand their menu, prep times, and constraints deeply so you can design a service around them, not impose one on them.
- 4.Use organic word-of-mouth as your north star for early growth: if customers don't share your product with friends without incentives, you don't have product-market fit. Measure and optimize for this before spending on paid ads.
- 5.Define your market niche tightly (e.g., international students who want specific cuisines; office workers in a 2-block radius) and dominate it completely before expanding. This prevents scattered focus and gives you a repeatable model to replicate in new markets.
Similar Companies
Zoom
$12.0M/moZoom is a freemium SaaS video conferencing platform founded by Eric Yuan in July 2011 after he left Cisco to build a next-generation collaboration solution. The company has grown to 850,000+ paying customers across individual, SMB, and enterprise segments, generating over $12M in monthly recurring revenue with approximately 100% year-over-year growth. Rather than focusing on customer stickiness or aggressive growth targets, Zoom emphasizes customer happiness and organic word-of-mouth acquisition, which has proven highly effective in driving viral adoption.
Plunge
$10.0M/moPlunge is a hardware company that manufactures and sells at-home cold plunge devices. Founded in 2020 by Ryan Duey and Michael after their brick-and-mortar float therapy and sauna businesses were impacted by COVID, the company grew from $270k in first-year revenue to $120M+ ARR in four years. Their success is driven by influencer gifting, organic word-of-mouth, and highly efficient paid advertising (7-10x ROAS on Facebook and Google).
G2
$5.0M/moG2 is a leading business software review website and marketplace founded in 2012 by Godard Abel. The company has scaled to over 500 employees and raised $257 million in capital, achieving unicorn status at a $1.1 billion valuation. G2 generates over $5 million in MRR today and targets $100 million in ARR next year through its core G2 Marketing Solutions for vendors, plus complementary products like G2 Track (SaaS spend management) and G2 Deals (marketplace procurement).
Active Campaign
$4.2M/moActive Campaign started in 2003 as an on-premise email marketing solution built by Jason Vanderboom to fund his fine arts degree. After 10 years and 8 employees generating a couple million in revenue, he transitioned to a SaaS model starting at $9/month. The company now has over 60,000 customers generating over $50 million annually and employs 330 people, growing primarily through organic adoption, partnerships, and focus on the SMB market despite pressure to move upmarket.
NutriSense
$3.3M/moNutriSense is a direct-to-consumer metabolic health platform that pairs continuous glucose monitoring devices with proprietary software analytics and dietitian coaching. Launched in September 2019 with pre-sales in keto and Oura Ring Facebook groups, the company grew from under $1M MRR a year ago to $3.3M MRR today (3x growth), with 15,000-16,000 active paying customers and 170 employees. The business has raised $32M in funding across multiple rounds since a $250K seed in early 2020.