Latest / Marketplace Businesses with Fexingo: Two-Sided Networks, Liquidity, and Take Rates / How Marketplaces Profit From Search Depth
Transcript
- Lucas: If these conversations are useful for what you're building or running, we need to talk about something that rarely makes it onto the slide deck: search depth. Most marketplace operators are obsessed with the first transaction. They want you to book that flight or hire that freelancer as fast as possible. But the real margin expansion happens in the seconds after the click. Luna: Right. Because once they have your credit card on file, the friction drops to zero. So why aren't more platforms pushing for that second, third, or fourth item in the basket? Lucas: They are, but most do it clumsily. Look at Airbnb. For years, their take rate was pretty static around ten percent on the listing fee. But recently, they started aggressively promoting experiences, add-on services like professional photography, and even travel insurance during the checkout flow. It wasn't a new product launch; it was deeper integration into an existing search result. Luna: That makes sense. If I'm already looking at a cabin in Vermont, showing me a guided foraging tour right next to the calendar is less intrusive than sending me a separate email three days later. Lucas: Exactly. It's contextual relevance versus cold outreach. The key metric here isn't gross bookings anymore. It's average order value per active user, specifically driven by ancillary services. When Airbnb added those experiences, they didn't just increase fees. They increased the time spent on the platform and the number of touchpoints per trip. Luna: But doesn't that risk annoying the supply side? If hosts are getting bogged down managing extra service requests, does the quality of the core offering suffer? Lucas: It can, which is why the best platforms treat these ancillaries as managed services rather than open marketplaces. Airbnb controls the pricing and the cancellation policy for those experiences. They absorb the customer service burden so the host doesn't have to. That keeps the core inventory clean while extracting higher margins from the add-ons. Luna: So they're effectively using the credibility of the main brand to sell high-margin peripheral goods. It's like Amazon recommending a case for the Kindle you just bought. Lucas: Precisely. And look at Fiverr. They used to be purely about hiring a freelancer for a specific task. Now, when you hire a logo designer, the platform suggests branding packages, social media templates, or even domain registration. These are vertical integrations that lock the user into a broader ecosystem before they've even left the site. Luna: I love that example because it shows how search depth changes the definition of a 'task.' A logo isn't just a graphic; it's part of a brand identity workflow. Capturing that workflow is worth way more than the one-off fee. Lucas: The data supports this. In Q3 of last year, several major marketplaces reported that over thirty percent of their total revenue now comes from services outside their core category. That shift didn't come from acquiring new users. It came from mining the existing user base more efficiently. Luna: Thirty percent is huge. But how do they decide which ancillaries to push? You can't just throw everything at the wall. Lucas: They use predictive analytics based on search intent. If a user searches for 'luxury wedding planner,' the system knows they're likely also interested in floral arrangements, venue decoration, or even photography retouching. It's not random cross-selling. It's inferred needs based on the primary query. Luna: So the algorithm becomes a consultant. It's not just matching supply and demand; it's anticipating the full scope of the project. Lucas: Yes. And this requires a massive investment in data infrastructure. You need to know not just who is buying, but what else they might need. Platforms like Upwork have been doing this for freelancers for years. When you hire a developer, they suggest QA testing, DevOps setup, or cloud hosting referrals. Luna: That's interesting because those suggestions often come from verified professionals within the network. It feels less like an ad and more like advice from a peer. Lucas: That's the golden ticket. Trust transfers across categories. If I trust the platform to find me a good lawyer, I'm more likely to trust their suggestion for a related accounting service. The moat isn't just the network; it's the reputation spillover between adjacent markets. Luna: But there's a limit. At some point, the interface gets cluttered. If every search result has five upsells, does the user just ignore them all? Lucas: Good point. That's why timing matters. The best implementations happen post-conversion or in the sidebar, never blocking the primary action. Uber Eats does this well. After you order dinner, they immediately show you dessert options or drinks. It's sequential, not simultaneous. Luna: Sequential. That's a crucial distinction. It respects the user's immediate goal before introducing new ones. Lucas: Exactly. It's about respecting the journey. If you force a cross-sell too early, you kill conversion. If you wait too long, the impulse fades. The sweet spot is right after the commitment is made, when the user is still in a 'spending mode' mindset. Luna: So the operator's job shifts from acquisition specialist to retention architect. You stop chasing new faces and start nurturing the relationships you already have. Lucas: That's the pivot. And it's cheaper too. Acquiring a new customer in today's market can cost fifty dollars or more in marketing spend. Upselling an existing customer costs almost nothing in terms of CAC, though it does require better product design. Luna: It's basically turning the customer base into an asset that compounds value over time, rather than a churn metric you constantly fight against. Lucas: Right. Think about LinkedIn Premium. They don't just sell you access to messages. They sell you insights on who viewed your profile, advanced search filters, and learning courses. Each layer adds stickiness and increases the monthly recurring revenue without changing the core product. Luna: It's a subscription inside a subscription. Very clever. But does it work for one-time buyers? Like someone buying a house? Lucas: Zillow and Redfin are trying. They offer mortgage calculators, moving services, and home warranty referrals. It's harder because the frequency is lower, but the transaction size is bigger. Even a small percentage conversion on those ancillaries yields massive absolute dollars. Luna: High stakes, low frequency. The strategy has to be different. Less about impulse, more about utility and peace of mind. Lucas: Exactly. Peace of mind is a premium product. People will pay extra to feel secure in a large transaction. Marketplaces that can bundle that security seamlessly win the trust war. Luna: This brings us back to the idea of liquidity. Deeper search means more options available at every step, keeping the user engaged longer. It's liquidity multiplied by time. Lucas: Liquidity times engagement equals monetization. It's a simple equation that few operators execute perfectly. Most try to scale breadth first. Depth is harder because it requires intimate knowledge of the customer's lifecycle. Luna: So the winners won't be the ones with the biggest catalogs. They'll be the ones with the smartest interfaces. Lucas: Smartest interfaces that anticipate needs. It's proactive commerce rather than reactive search. That's where the margin is hiding. Luna: A couple of dollars a month is genuinely what keeps these going. Buy me a coffee dot com slash fexingo if you've gotten something out of these breakdowns. It helps us keep digging into the mechanics behind the markets. Lucas: Thanks for the support. It lets us stay independent and focus on the details that matter. Back to the topic, though. Luna: One last thought on this. If a marketplace fails to deepen its search, does it become vulnerable to competitors who do? Lucas: Absolutely. If a competitor offers a seamless bundle that saves you time and money, you'll switch. The lock-in effect of search depth is stronger than any single feature. It creates a habit loop that is very hard to break. Luna: So the moat isn't just data. It's the convenience of having everything in one place, orchestrated intelligently. Lucas: Convenience is the ultimate currency. And as algorithms get better at predicting what we want next, that convenience becomes increasingly valuable. The question for operators is whether they are building for the next click, or the next relationship.