Latest / Marketplace Businesses with Fexingo: Two-Sided Networks, Liquidity, and Take Rates / How Marketplaces Set Fees Without Killing Liquidity
Transcript
- Lucas: So, Luna, we've spent over a hundred episodes talking about how marketplaces raise their take rate. But there's a flip side that we haven't really dug into: what happens when the take rate is too high, and the marketplace starts to lose the liquidity it was built on? Luna: Right, we've always looked at it from the platform's perspective. What about the seller's breaking point? Lucas: Exactly. And there's a really interesting case from 2019 that I think frames this perfectly. Etsy raised its transaction fee from five percent to six and a half percent. In one move, they took their cut from sixty-five dollars on a thousand-dollar sale to ninety dollars. Sellers revolted. Some left. But Etsy's revenue jumped. Luna: So did Etsy actually cross a line, or did they just get away with it? Lucas: That's the question. And the answer involves what I like to call the take-rate cliff. Every marketplace has a threshold where the fee becomes high enough that sellers start to seek alternatives, or off-platform transactions become too tempting. The trick is finding that cliff without falling off it. Luna: And in Etsy's case, they seem to have stepped right up to the edge. Their gross merchandise sales are still growing, even after the fee hike. Lucas: Right. And that's partly because Etsy is a specialized marketplace. A handmade ceramicist doesn't have a lot of other platforms with the same buyer traffic. So the take-rate elasticity is lower. But for more commoditized goods, the picture changes. Lucas: Think about eBay. Sellers there have been complaining for years about fee creep. And eBay's take rate is around twelve to fifteen percent when you factor in listing fees and payment processing. That's nearly double Etsy's. Luna: But eBay still has massive liquidity. Why don't sellers just pack up and go to another platform? Lucas: Because liquidity begets liquidity. A seller might pay fifteen percent on eBay, but if that's where the buyers are, the cost is worth it. The switching cost isn't just the fee—it's the loss of access to demand. Luna: So you're saying the take rate is almost like a tax on access to demand. And if the demand is strong enough, sellers will pay almost any price. Lucas: Almost. But there is a ceiling. And some marketplaces have learned that the hard way. One example that often gets cited is the early days of Uber Eats. They took massive subsidies to build supply on both sides, then started raising the commission they charged restaurants from fifteen percent to thirty percent. A lot of independent restaurants pushed back, and some actually started directing customers to order through their own channels to avoid the fee. Luna: So the take-rate cliff isn't just about the fee itself—it's about how the fee compares to the alternatives. If a restaurant can get a customer to order directly, they avoid the fee entirely. Lucas: Exactly. And that's why marketplaces have to constantly justify their take rate. They're not just charging for a transaction; they're charging for the discovery, the trust, the payment processing, the support. If the seller feels like they can replicate those benefits on their own, the fee becomes harder to justify. Luna: And that's where the psychology of fee anchoring comes in. I read somewhere that how you present the fee can matter as much as the fee itself. Lucas: Absolutely. There's a classic example from Airbnb. Hosts pay a service fee of about three percent, and guests pay a service fee that can be up to fourteen point two percent. But Airbnb frames the guest fee as a 'service fee' that's separate from the nightly rate. So when a host compares their three percent to Etsy's six and a half percent, they feel like they're getting a good deal. But the total take rate that Airbnb earns across both sides is much higher than the host fee alone. Luna: So they're effectively splitting the take rate between two sides, which makes each side feel like they're paying less. That's clever, but doesn't it risk a backlash if people figure it out? Lucas: It can. And that's exactly what happened with Ticketmaster when they introduced dynamic pricing. The take rate wasn't the issue as much as the transparency. When consumers feel like the fee is hidden or unjustified, they push back. So the framing of the fee is a big part of maintaining liquidity. Luna: So the take rate is not just a number—it's a story you tell to both sides of the market. Lucas: Exactly. And that story has to be consistent with the value you're delivering. If you're charging a premium take rate, you need to deliver premium value. That's why some marketplaces, like StockX, can justify a higher take rate because they're offering authentication and a transparent market. Lucas: But the real art is in knowing when to lower the take rate to grow liquidity. A lot of marketplaces will actually subsidize the supply side early on to attract sellers, and then gradually raise the take rate once the network is established. That's what I call the 'liquidity-first' approach. Luna: And we've talked about that in the context of cold-start problems, but how does that play out in the long run? Lucas: Well, take Airbnb again. In their early days, they didn't charge hosts anything. They only charged guests a small fee. That helped them attract hosts, who then brought guests. Once they had both sides, they introduced the host fee. But they kept it relatively low to avoid disrupting the equilibrium. Luna: So the take rate is like a dial that you turn up as you gain more liquidity, but you have to be careful not to turn it too fast. Lucas: Right. And sometimes you even have to lower it. There have been cases where marketplaces have had to cut their take rate to retain sellers. For instance, when Instacart faced competition from DoorDash, they reduced their commission rates for some grocers to stay competitive. The take rate became a strategic weapon, not just a revenue source. Luna: That makes me think about the concept of price elasticity in marketplaces. The take rate sensitivity varies depending on the type of seller and the type of buyer. Lucas: Exactly. A professional seller who relies on the marketplace for their entire income is going to be very sensitive to fee changes. But a casual seller who's just cleaning out their closet on eBay might not even notice a one percent difference. Luna: So the take-rate cliff isn't a single point—it's a range that depends on the seller's alternatives and their dependence on the platform. Lucas: And that's why data is so important. Marketplaces with good data can segment their sellers and adjust fees dynamically, almost like surge pricing, to maximize liquidity while still capturing value. Lucas: We've seen some marketplaces experiment with tiered fee structures—lower fees for high-volume sellers, higher fees for low-activity ones. That's a way to keep the most valuable suppliers happy while still earning the same overall take rate. Luna: That sounds like a smart approach. But isn't there a risk of alienating smaller sellers who feel like they're being penalized for being small? Lucas: There is. And that's where the trade-off becomes really tricky. A marketplace like Etsy is built on small sellers, so they have to be careful. That's why they're always framing their fee structure as 'supporting the community.' But at the end of the day, it's a business. Luna: So what's the takeaway for someone who's building a marketplace today? How do you set your take rate from day one? Lucas: I'd say start lower than you think you need to. It's much easier to raise fees later once you've proven your value than to lower them after a backlash. And keep a close eye on your liquidity metrics—if you see a dip in seller activity after a fee change, you might have hit the cliff. Lucas: And remember that the take rate is not just a pricing decision—it's a product decision. You have to continuously earn that fee by delivering more value to both sides. Luna: That's a great place to leave it. Thanks for breaking down the take-rate cliff, Lucas. Lucas: Anytime, Luna. And if you're building a marketplace, don't be afraid to experiment with your take rate, but always measure the impact on liquidity.