Latest / Marketplace Businesses with Fexingo: Two-Sided Networks, Liquidity, and Take Rates / How the Best Marketplaces Use Loss Leaders to Bootstrap Liquidity
Transcript
- Lucas: So there's this moment every marketplace founder hits — you've built the platform, the code works, the design is clean, and then you sit there with zero transactions. No supply because there's no demand. No demand because there's no supply. Classic chicken and egg. Luna: Right, the cold-start problem. And the most muscular solution to it — the one that's been used by some of the biggest marketplaces we know — is the loss leader. You deliberately lose money on one side of the transaction to get the flywheel spinning. Lucas: Exactly. And I want to be specific about what a loss leader actually means in a marketplace context, because people sometimes throw the term around loosely. It's not just 'spend money on marketing.' It's where you price a transaction below your cost — on purpose — to acquire a user who then becomes valuable later. Luna: The classic example has to be Uber in the early years. They were essentially paying for both sides of the ride. Riders got a heavily subsidized fare — sometimes a ride that cost Uber twenty dollars would be charged to the rider for five. And drivers still got paid their full fare. Lucas: Right. Uber was losing money on every ride, but they were buying a behavior. They were training riders that opening the app and getting a car in five minutes was normal, and they were training drivers that the app would deliver a steady stream of requests. The loss was a deliberate investment in habit formation. Luna: And the bet was that once both sides were hooked, you could gradually pull back the subsidies. Which they did — over years, not months. But the initial capital required was enormous. That's why this strategy really only works if you have venture funding that's willing to tolerate deep losses for a long time. Lucas: That's the key tension. A loss-leader strategy is a liquidity-bootstrapping tactic. It's not a business model. The question every marketplace operator has to answer is: can we eventually make money on these users, or are they only here because we're paying them to be here? Luna: Yeah, and some marketplaces never manage the transition. I think about home-services platforms like Thumbtack in its early days. They were essentially paying pros to submit quotes — giving them credits or cash for each response they sent to a customer. The pros loved it, but the moment Thumbtack pulled back the subsidies, a lot of them left. Lucas: Because the pros weren't there because Thumbtack was the best place to find clients. They were there because Thumbtack paid them to respond. The platform hadn't actually built enough organic demand yet. That's the danger — you can end up subsidizing a fake marketplace that collapses when the money stops. Luna: So when does a loss leader actually work? What's the difference between Uber's playbook and Thumbtack's? Lucas: I think it comes down to two things. First, the subsidy has to be on the side that has more elastic behavior — the side you can actually change habits on. Uber subsidized riders, not drivers, because a rider's decision to open an app is more habit-driven. A driver's decision to turn on the app is more economic — they go where the money is. Luna: Interesting. So Uber subsidized the side that had the potential for long-term retention, not the side that would just chase the highest payout. Lucas: Right. Second, you need a plan to remove the subsidy that doesn't destroy the experience. Uber did it by getting the unit economics to improve through efficiency — shorter wait times, better routing, higher driver density — so that even at full price, the rider perceived more value than a taxi. The value prop shifted from 'cheap' to 'convenient.' Luna: Let's talk about another approach — the one where the loss leader is on the supply side, but it's a capital asset, not a cash subsidy. OpenTable famously gave free reservation tablets to restaurants. Those tablets cost them something like three to five hundred dollars each. They were essentially paying restaurants to join the platform. Lucas: That's a great example because the loss is upfront hardware, not per-transaction. OpenTable's bet was that once a restaurant had the tablet on the host stand, they'd never go back to a paper book. And the switching cost was enormous — not just the tablet itself, but the training, the habit, the integration with the restaurant's workflow. So the acquisition cost was high, but the churn was incredibly low. Luna: And on the demand side, diners didn't need a subsidy — they wanted the convenience of booking online. So OpenTable only had to lose money on one side, and the other side came naturally. That's the ideal scenario, isn't it? When one side is inherently motivated and the other needs a nudge. Lucas: Exactly. The best marketplace loss leaders target the side that has the highest friction to adoption but also the highest stickiness once adopted. Restaurants had never used digital reservation systems — the tablet was a huge leap. But once they did, they were locked in. Riders had never used a car-hailing app — once they did, the convenience was addictive. Luna: Of course, this whole approach assumes you have the capital to sustain the losses. And in 2026, the funding environment is different from the ZIRP era when Uber raised billions. VCs are much more focused on unit economics from day one. Can you even run a loss-leader strategy today? Lucas: You can, but you have to be much more surgical about it. Instead of subsidizing every transaction, you might subsidize the first three transactions and then taper off. You might cap the total subsidy per user. You might use a loss leader only in specific geographies where you need to break a dominant competitor's hold. The principle is the same — buy the behavior, then prove the unit economics before scaling the subsidy. Luna: So it's less 'spray and pray' and more 'test and invest.' Lucas: Exactly. And if these conversations about marketplace strategy are useful for what you're building or running, we want to keep them coming without any ad breaks. That's a deliberate choice — no sponsors, no mid-roll interruptions. If you find value in that approach, the simplest way to support it is at buy me a coffee dot com slash fexingo. That's it. No tiers, no perks, just helping us stay independent. Luna: Yeah, and it genuinely makes a difference. We hear from listeners who say they appreciate the ad-free format. If that's you, the link is buy me a coffee dot com slash fexingo. Lucas: Alright, back to marketplaces. So we've talked about subsidizing demand and subsidizing supply with hardware. But there's a third loss-leader model that's interesting — subsidizing the platform's own cost structure to offer a lower take rate initially. Luna: Meaning you charge less commission than you need to break even, hoping to make it up on volume later? Lucas: Right. A lot of marketplace SaaS companies do this — they offer the first three months free or at a steep discount. But the risk is that you attract merchants who are purely price-sensitive and will leave the moment you raise the rate. The signal you want is that your platform provides enough value that merchants stay even at a higher take rate. Luna: So the loss leader has to be paired with a product that actually delivers increasing value over time. If your only moat is the subsidy, you're in trouble. Lucas: Exactly. And that's why the most successful marketplace loss leaders aren't just about pricing — they're about building dependencies. Uber built dependency on convenience. OpenTable built dependency on workflow. Etsy, in its early days, essentially gave free listings to sellers, which was a loss leader, but they also built a community and a buyer audience that sellers couldn't replicate on their own. Luna: So the loss leader is really a loan against future loyalty. You're spending cash today to earn a relationship that you hope will pay dividends tomorrow. And the marketplace that does it best is the one that understands which side of the platform that loan actually builds a lasting asset on. Lucas: That's a great way to put it. And I think the smartest move most marketplaces make is to start phasing out the loss leader before the users expect it — while the value is still going up. If you wait until you have to raise prices because your margins are collapsing, you've lost the narrative. But if you raise prices while still improving the product, users grumble but they stay. Luna: It's like a subscription price increase — the worst time to do it is when you're about to report bad earnings. You want to do it when the goodwill is high. Lucas: Exactly. So for anyone building a marketplace today, I'd say: use a loss leader if you have to, but have a clear thesis on which behavior you're buying and how you'll wean users off it. And if you're pitching investors, be ready to show that your unit economics improve by, say, thirty percent as a cohort matures. That's the proof that the subsidy was an investment, not a handout. Luna: And maybe the most honest question a founder can ask is: if I stopped the subsidy today, would my marketplace still have liquidity? If the answer is no, you haven't built a marketplace — you've built a promotion. Lucas: That's the line. A loss leader is a tool, not a strategy. Use it wisely, and you bootstrap a network that can stand on its own. Misuse it, and you're just burning cash on a rented marketplace.