Grant LaFontaine founded Whatnot, a live shopping marketplace now valued at $20 billion, growing from a niche Funko Pop app to a platform where millions of sellers — including individuals who built nine-figure businesses — sell everything from collectibles and fashion to fresh seafood and gold bars.
Whatnot today: scale and scope
Whatnot is a live video shopping platform where sellers stream auctions and sales in real time; buyers watch, chat, and purchase instantly.
The platform hosts millions of sellers and buyers across categories including collectibles (Funko Pops, sports cards, comics), fashion, sneakers, electronics, fresh food (crabs, meat), and precious metals.
As of the interview: ~1,400 employees, $1.5B+ raised, multiple sellers doing $100M+ annual GMV, and total GMV exceeding $1B in 2022 (up from $2.3M in 2020).
Grant estimates live commerce is currently low single-digit percentage of US e-commerce but will reach 30%+ within a decade, mirroring China’s ~40% penetration — implying a potential trillion-dollar market outside Asia.
Origin story: from Tokyo bar to Y Combinator
Grant and co-founder Logan met while working in consumer internet (Grant at Facebook/YouTube, Logan running engineering at resale marketplaces); both had prior startup experience.
In late 2019, drinking in a Tokyo bar, they bought 50 domains intending to build a marketplace — initially a “better Craigslist” with delivery logistics.
Grant cold-called delivery executives (Bob’s Discount Furniture, etc.) on LinkedIn; within 24 hours realized unit economics made no sense for low-value used goods.
Pivoted to collectibles after analyzing eBay category data (via Terapeak) showing explosive growth in Funko Pops and Pokémon; Funko had more volume than comic books.
Applied to Y Combinator with a working app built in a week; accepted despite being a “Funko Pop marketplace” — investors later passed on the seed round for the same reason.
Solving the cold-start problem: creative seller and buyer acquisition
Classic marketplace chicken-and-egg: no buyers without sellers, no sellers without buyers. Whatnot solved for demand first by acting as the sole seller.
Grant and a team in Brazil sourced Funko Pops from 100+ online retailers, authenticated them, and resold on Whatnot; built a pricing algorithm scraping competitor sites and using liquidity floors (e.g., 10 sales/30 days) to avoid unsecurable inventory.
Buyer acquisition: (1) partnered with Funko Pop YouTube influencers — Grant cold-called dozens in December 2019; (2) built a viral “Grails giveaway” mechanic — weekly raffles for $500–$1,000 chase items where each share/referral earned extra tickets, taking over Funko subreddits and Facebook groups.
First giveaway flopped (104 entries, zero purchases); second ~2.5x; exponential growth followed as the user base compounded.
Opened to external sellers in late February 2020 once demand existed; cross-listed seller inventory on eBay via Whatnot’s account to guarantee sales (Airbnb-style Craigslist hack).
The live shopping inflection point: July 2020
After YC demo day (March 2020: $25K GMV), seed round in Aug/Sept 2020 ($300K GMV), Grant launched the first live auctions feature in July 2020.
Live transformed the experience: sellers could run real-time shows, buyers discovered products socially, engagement and conversion skyrocketed.
Growth went to 100%+ month-over-month; December 2020 closed at $2.3M GMV with ~18 employees.
COVID accelerated adoption: comic shops and brick-and-mortar sellers moved online, finding larger audiences than physical foot traffic allowed.
Rejects “visionary founder” narrative: “We’re not a vision-driven company. We’re a user-driven company.” Vision matters for direction, but customer demand is the ultimate arbiter.
Stories are often “a mechanism to convince people to follow you versus something of real substance”; real substance has depth.
Example: investors passed on Whatnot because “Funko Pops are a small market” — but every great consumer company starts niche (eBay: broken laser pointers; PayPal: eBay seller tool) and expands.
Avoids hiring big-company executives: “Someone needs to know and articulate in specific detail what they’re doing and how they’re doing it.” Pedigree ≠ competence in early-stage chaos.
Core discipline: “Seek the truth” — go to the root of problems, demand simple explanations, reject jargon that obscures understanding.
Example: discovery algorithm regression — team cited A/B tests showing engagement up; Grant forced deep dive into seller-level elasticity curves, revealing that shifting impressions from high-elasticity sellers shrank total GMV despite higher buyer engagement. A/B tests pollute both sides in network systems.
Management evolution: operating system and culture
Grant’s “operating system”: how to run the company at scale — operating rhythm, decision-making, accountability, execution speed.
Early culture: blunt feedback (“this sucks, fix it”); as company grew, had to soften delivery to avoid intimidation while maintaining rigor.
“If you think you’re moving fast, you’re probably not. If you have to talk about it, you’re moving too slow.”
Hires executives with explicit contract: “Your methods are at best neutral when you come in.” Culture rewards getting to the right answer, not hierarchy.
Long-tenured leadership team (4+ years) creates shared language and permission to challenge deeply.
Future outlook and ecosystem opportunities
Live commerce will be 30%+ of e-commerce in 10 years; top sellers will do $1B+ GMV (as in China).
MCNs (multi-channel networks) — agencies helping sellers run live shopping operations.
Tools/services for seller operations (inventory, logistics, analytics, streaming production).
Grant still collects (recently: Type 1 original photographs) and plays hockey; emphasizes family accountability and investor accountability over personal celebration.
