Latest / Marketplace Businesses with Fexingo: Two-Sided Networks, Liquidity, and Take Rates / How Marketplaces Use Freemium to Solve Liquidity
Transcript
- Lucas: Marketplaces have a fundamental problem at birth: no buyers without sellers, no sellers without buyers. The classic solution is to subsidize one side — Uber paid drivers to sit idle, Airbnb photographed listings for free. But a quieter, more elegant approach has been gaining traction. Give away the core value for free, then charge only when a transaction happens. Luna: Freemium. It's not a new concept — Dropbox did it beautifully — but it's surprisingly rare in pure marketplace contexts. Most people think of freemium as a software play. Lucas: Right. And that's exactly why it's interesting. When a marketplace can deliver value without a transaction — a tool, a profile, a preview — the free tier becomes a liquidity engine. Let's start with the canonical example: Dropbox isn't a marketplace per se, but it shows the mechanism perfectly. Luna: Dropbox gave away two gigabytes of storage for free. No credit card, no time limit. The referral program then rewarded both sides — referrer and referee — with extra space. That created a viral loop. Lucas: Exactly. At its peak, that referral program drove 35 percent of all daily sign-ups. Dropbox turned every free user into a salesperson. Now, fast-forward to a true marketplace: Upwork. Freelancers create profiles, submit proposals, and message clients — all for free. The platform only takes a cut — about 20 percent on the first $500 with a client — when work is actually paid. Luna: That's a clever liquidity hack. By removing the upfront cost for freelancers, Upwork solves the cold start on the supply side. The risk is entirely on the platform: they host profiles and facilitate matches with zero guarantee of revenue. Lucas: And it works because the marginal cost of hosting a profile is near zero, while the potential lifetime value of a successful freelancer is high. Upwork's take rate on total billings has hovered around 20 percent on their managed marketplace, and they've been profitable on an adjusted basis. Luna: But not every marketplace can pull this off. Uber can't let drivers offer free rides. Airbnb can't give away a free night's stay. The service itself has a cost. Lucas: That's the key constraint. Freemium works when the platform can deliver a valuable slice of the experience without a transaction. Canva is a brilliant example. It started as a free design tool — no marketplace at all. They amassed 60 million users by offering templates and drag and drop design for free. Then they opened a marketplace for premium templates, photos, and fonts. Luna: So the free tier built the demand side. When Canva introduced transactions, the buyers were already there. The marketplace didn't have to bootstrap. Lucas: Precisely. Canva takes a 40 percent cut on marketplace sales, and their revenue hit $1.2 billion in 2024. The free tier essentially pre-solved liquidity. Compare that to a marketplace like Etsy, which had to build buyer traffic through SEO and advertising from scratch. Luna: There's a behavioral economics layer too. The 'zero price effect' — people over-index on free. Dan Ariely's experiments showed that when a price drops from one cent to zero, demand jumps disproportionately. Marketplaces that can offer a free tier tap into that irrational excitement. Lucas: That's real. Dropbox could have charged $2 for the basic tier and probably killed the viral loop. The psychology of 'free' is a liquidity catalyst that paid acquisition can't match. But there's a danger: if the free tier is too good, nobody pays. Upwork faces this — many freelancers use the free tier forever, and Upwork only makes money when they land a gig. Luna: So the freemium marketplace has to carefully design what's free and what's paid. The free tier should be valuable enough to attract users, but leave a clear reason to upgrade. In Upwork's case, the reason to pay is 'getting paid' — the transaction itself triggers the fee. Lucas: Another risk is cannibalization. If a marketplace relies on listing fees or subscriptions, a free tier could eat those revenue streams. Look at eBay: they charge insertion fees for listings. If they made listings free, they'd lose billions. But for Upwork, the transaction fee model aligns perfectly with freemium — zero marginal cost for the free tier, revenue only when value is created. Luna: Let's talk about a marketplace that tried freemium and struggled. Fiverr originally let sellers create gigs for free, but buyers had to pay. The model worked, but they found that too many low-quality free gigs cluttered the platform. Lucas: Quality control is a real issue. Freemium lowers the barrier to entry, which can flood the supply side with poor offerings. Fiverr had to introduce a 'Fiverr Pro' tier and stricter vetting. The freemium model works best when the platform can maintain quality standards algorithmically or through reputation systems. Luna: Speaking of reputation, Stack Overflow is a fascinating case. It's a marketplace for knowledge — questions and answers. The core service is free. They make money through job listings and advertising. The free Q&A creates an enormous audience that they then monetize on the talent side. Lucas: That's a two-sided marketplace where the free side is so valuable it becomes the moat. No competitor could replicate 10 million answered questions. Stack Overflow's take rate on job listings is about 20 percent. The free tier is their biggest competitive advantage. Luna: So the pattern emerging is: freemium works best when the free tier itself generates a network effect — more free users attract more free content, which attracts more free users, and the paid side harvests that attention or matches. Lucas: Exactly. LinkedIn is another one. Free profiles, free networking, free content. Then they charge for premium features like InMail and job posting. The free network is the asset. LinkedIn's talent solutions revenue, which includes job listings and recruiter subscriptions, was over $7 billion in 2024. The free tier built the largest professional graph on earth. Luna: But LinkedIn also has a paid subscription for users — Premium. So it's a blend. The free tier is generous, but not unlimited. They limit who you can message, how many search results you see. That creates a conversion funnel. Lucas: The takeaway for marketplace builders: if you can deliver a valuable core experience at near-zero marginal cost, freemium can be your fastest path to liquidity. But you need to know exactly what your paid trigger is. For Upwork, it's the transaction. For Canva, it's the premium asset. For LinkedIn, it's the connection and job search. Luna: And you need to measure the conversion rate from free to paid. If it's too low, you're just burning server costs. Upwork's conversion from free user to paying client is estimated around 5-10 percent, but the lifetime value of those clients makes it worthwhile. Lucas: Right. The metrics that matter are not just sign-ups, but active free users who eventually transact. You want to track the 'time to first paid action' and the 'free-to-paid conversion rate'. If those are healthy, freemium is a liquidity machine. Luna: I want to bring this back to a more recent example: Substack. They let writers publish newsletters for free. Only when a writer chooses to charge — and readers choose to pay — does Substack take a 10 percent cut. The free tier is the entire writing platform. Lucas: Substack's freemium is almost pure Upwork model. The platform provides the tool for free, builds the network, and monetizes only the transactions that succeed. It's a marketplace for attention that becomes a marketplace for money. As of early 2026, Substack had over 3 million paid subscriptions. The free publishing tier was the launchpad. Luna: If these conversations are useful for what you're building or running, honestly, if today was worth a coffee to you, that's the link — buy me a coffee dot com slash fexingo. It keeps the show ad-free and independent. Lucas: Yeah, it helps us keep digging into models like this. So back to the mechanics: one challenge with freemium is that you're essentially running a charity for free users until they convert. You need a healthy balance sheet or patient investors. Luna: Or you need a very low cost to serve free users. Cloud storage costs Dropbox a lot per user — about $0.01 per gigabyte per month — but the referral viral loop brought in users so cheaply that the unit economics worked. Lucas: Let's synthesize. Freemium is not a one-size-fits-all. It works when the free tier is a genuine value engine, not just a trial. It works when the marginal cost is low. And it works when the paid trigger is clear and frictionless. Marketplaces that can check those boxes have a powerful tool to solve the chicken and egg problem. Luna: I think the biggest mistake I see is marketplaces treating freemium as a marketing channel instead of a product strategy. The free tier has to be a product in itself — one that users love and that naturally leads to a paid moment. Lucas: That's a great way to put it. And for next episode, I want to explore what happens when marketplaces flip the model — charging users for access instead of per transaction. Subscription marketplaces. That's a whole different liquidity dynamic. Luna: Looking forward to it. For now, if this episode sparked an idea, that coffee link is still open.