Latest / Marketplace Businesses with Fexingo: Two-Sided Networks, Liquidity, and Take Rates / Why the Best Marketplaces Use Subscription Over Per-Transaction Fees
Transcript
- Lucas: If you run a marketplace, the default answer to 'how do we make money?' is usually 'take a cut of every transaction.' Fifteen percent, twenty percent — whatever the market will bear. But a growing number of the most durable marketplaces have quietly moved away from that model toward something much simpler: a flat subscription fee. Luna: Right, and it's not just Amazon Prime. We're seeing it on both sides of the marketplace — charging buyers, charging sellers, sometimes both. Lucas: Exactly. Let's start with Amazon Prime because it's the biggest example. Over two hundred million subscribers globally, each paying roughly one hundred and thirty-nine dollars a year in the US. But here's the thing: Prime isn't really a 'marketplace fee' in the traditional sense. It's a membership that unlocks free shipping, video, music, and — crucially — exclusive deals. Amazon figured out something early: if a customer has already paid an annual fee, they're going to shop more to justify that cost. The subscription actually increases transaction volume. Luna: So the take rate isn't on each sale — it's upfront. But that changes the relationship entirely. The buyer is now incentivized to consolidate their shopping on Amazon. Lucas: And the seller benefits too, even though they're not paying the subscription. Sellers on Amazon get access to a pool of buyers who have already committed. Prime members spend about two and a half times more annually than non-Prime members. That's not accidental — the subscription creates a psychological sunk cost that drives repeat behavior. Luna: What about marketplaces where the subscription is charged to the seller side? I'm thinking of Care.com — the platform for finding babysitters, senior caregivers, pet sitters. Lucas: Great example. Care.com actually charges families a monthly membership fee — roughly thirty-five to forty-five dollars a month — to message caregivers. They don't take a per-booking cut. The caregivers themselves can create profiles for free. So the monetization is entirely on the demand side, and it's a fixed fee regardless of how many bookings you make. Luna: And this solves a specific problem in caregiving marketplaces. If you charged per transaction, families might hesitate to message multiple candidates because each message costs something. With a subscription, they can reach out to ten, fifteen potential caregivers without incremental cost — and that improves the match quality. Lucas: Exactly. The subscription eliminates friction on the discovery side. Care.com's data suggests that families who subscribe end up hiring faster and with higher satisfaction because they're not rationing their outreach. And for caregivers, they get more inquiries because families aren't nickel and dimed per connection. Luna: But there's a downside, right? If you charge a flat fee, you risk adverse selection. The families who sign up might be the ones who plan to use the service heavily, while light users churn out. Lucas: That's the classic threshold problem. If your subscription is forty dollars a month, someone who only needs a babysitter once a quarter will do the math and decide it's not worth it. So you have to either offer a lower tier — like a pay per booking option — or make sure the value is so obvious that even occasional users feel the subscription pays for itself. Luna: Some marketplaces solve that by offering both. I think of Angi — formerly Angie's List. You can pay per project or buy a membership that gives you discounted rates and priority booking. They segment the market. Lucas: Right, and that hybrid model is actually becoming more common. But the pure subscription approach works best when the transaction frequency is naturally high. Amazon works because people buy stuff every week. Care.com works because families typically need recurring care — weekly, daily even. If your marketplace is for weddings or vacation homes, subscription is a much harder sell. Luna: Let's talk about the other side: marketplaces that charge sellers a subscription. Etsy, for example, charges a listing fee and a transaction fee, but they also offer Etsy Plus — a subscription for sellers that gives them credits for listings, pattern websites, and discounted tools. Lucas: And that's smart because it aligns incentives. A seller who subscribes to Etsy Plus is likely more committed to the platform. They'll optimize their listings, respond faster, ship on time — because they've invested. That improves the buyer experience, which in turn draws more buyers, which makes the subscription even more valuable. It's a virtuous cycle. Luna: But there's a risk. If you charge sellers a subscription, you might discourage the long tail of casual sellers who are just testing the platform. That's why Etsy keeps the basic listing model — low fixed cost per listing — and offers the subscription as an upgrade. Lucas: Which brings us to a key principle: the subscription should fund a better experience, not just be a rent extraction. Amazon Prime funds faster shipping and video content. Care.com's fee funds background checks and customer support. Etsy Plus gives sellers tools to improve their shops. If the subscription doesn't buy something tangible, customers will churn. Luna: And that's where many marketplaces fail. They think subscription is just a more predictable revenue stream than transaction fees. But the best ones treat it as a product feature that strengthens network effects. Lucas: Let's zoom out. The reason subscription models are gaining traction is that they reduce the friction of repeated transactions. In a per-transaction model, the marketplace is constantly re-proving its value. Every single time a buyer buys, they see a line item 'service fee' and think, 'Why am I paying this?' With a subscription, that mental accounting disappears. The fee is already paid. The marginal cost of another transaction is zero. And that drives exactly the kind of repeat behavior that makes a marketplace sticky. Luna: It's also easier to forecast revenue. If you know you have two hundred million Prime subscribers at one thirty-nine a year, that's about twenty-eight billion dollars in subscription revenue before you sell a single product. That's hugely valuable for planning and investment. Lucas: But let's be honest about the challenges. First, the upfront commitment scares off price-sensitive users. Second, you need to deliver value consistently — if a subscriber doesn't transact for three months, they start questioning the subscription. And third, the pricing has to be just right: too low and you leave money on the table, too high and you choke off growth. Luna: There's also the question of whether subscription models can work in developing markets where card penetration is lower. Amazon in India, for example, offers Prime at a much lower price — around seven hundred rupees a year, which is less than ten dollars — and they accept mobile wallet payments. Lucas: Exactly — locale matters. But the core insight holds: a subscription flips the incentive from 'maximize transactions' to 'maximize value per subscriber.' That's a healthier long-term orientation for a marketplace. You stop optimizing for conversion on a single purchase and start optimizing for lifetime customer experience. Luna: So if we're evaluating a marketplace startup today, what's the tell? How do you know if a subscription model is right for them? Lucas: I look at three things. One: transaction frequency. If the average user transacts less than once a quarter, subscription is tough. Two: switching costs. If users can easily take their business elsewhere, the subscription just adds friction and they'll leave. Three: value density. Can you bundle enough value — shipping, tools, exclusivity — to make the subscription feel like a steal? If you can tick all three, subscription is probably stronger than per-transaction. Luna: And what about marketplaces that have switched from per-transaction to subscription? Any success stories besides Amazon? Lucas: One that comes to mind is Reverb — the marketplace for musical instruments. They introduced Reverb Bazaar, a subscription for sellers that gives them discounted fees and promoted listings. It's still early, but early data suggests that subscribed sellers list more items and sell faster. The subscription seems to be aligning seller behavior with platform quality. Luna: That makes sense. Sellers who pay a flat fee want to maximize their exposure, so they list more and price competitively. Buyers see more selection and better prices, so they buy more. That's network effects on steroids. Lucas: Right. And the marketplace gets predictable revenue without having to increase take rates, which is a delicate balance. You can't just keep raising the transaction fee — sellers will leave. But a subscription that delivers real value? That's a different conversation. Luna: You know, this episode made me think about the business model of the show itself. We're providing useful analysis every week — no ads, no sponsors — and that's only possible because of listener support. Lucas: Yeah, exactly. If you found today's breakdown useful — the subscription versus per-transaction framework, the Prime and Care.com cases — buy me a coffee dot com slash fexingo is where you can keep this going. Just a simple, no-strings way to say 'this was worth my time.' We keep it ad-free that way. Luna: It's the same principle we talked about: a small upfront contribution unlocks ongoing value. In our case, that value is episodes like this one — and listeners seem to appreciate the direct relationship. Lucas: Exactly. So back to our framework — let's talk about one more case where subscription failed. There was a marketplace for freelance designers called Dribbble that experimented with a pro subscription for job seekers. It didn't work because the value wasn't tangible enough. Subscribers didn't see a clear return, so they churned. That's a cautionary tale. Luna: So the lesson is: don't slap a subscription on just because you want recurring revenue. You have to engineer the product around it. Lucas: Exactly. Subscription is a product decision, not a pricing decision. When it works, it aligns everyone — buyers, sellers, platform — around long-term value. When it's forced, it adds friction and destroys liquidity. Luna: Alright, so next time you're evaluating a marketplace, look at their pricing model. If they've moved to subscription, ask why. If they haven't, ask if they should. That's a solid framework. Lucas: And that's the kind of thinking that separates good operators from great ones. Thanks for listening, and we'll be back next week with another angle on marketplace mechanics.