Latest / The Platform Economy with Fexingo: Marketplaces, Networks, and Multi-Sided Businesses / How Marketplaces Build Trust Without Ratings
Transcript
- Lucas: Look at the last thing you booked on Airbnb or hired through a freelance platform. If you remember it clearly, it probably had a five-star rating attached to the profile. But here is the thing: that visible score is disappearing faster than most people realize. Luna: I noticed that too. I just booked a cleaner, and instead of seeing her average score out of five, I just saw a little badge that said 'Super Reliable.' It felt less like data and more like a label. Lucas: Exactly. That badge is what we are calling an opaque trust signal. Platforms are moving away from transparent, comparable metrics toward internal, non-comparable reputation layers. The goal isn't to help you compare suppliers better. It is to make comparison impossible so you rely on the platform's judgment. Luna: So they are removing the ruler so we stop measuring them against each other? Lucas: Precisely. For years, the industry standard was the star system because it was intuitive. But it created a massive gaming problem. On Uber, drivers realized that getting one bad ride could tank their income, so they became overly aggressive about pleasing passengers rather than doing the job safely. On Airbnb, hosts started leaving subtle clues in photos to bias reviewers before they even stayed. Luna: It turns the interaction into a performance review instead of a service exchange. Lucas: Right. And the data shows this is costly. When you look at marketplaces that switched to hidden scores, initial friction went up because users couldn't instantly verify quality. But over six months, dispute rates dropped by nearly fifteen percent because the feedback loop stopped being punitive and started being diagnostic. Luna: That seems counterintuitive. If I can't see the score, how do I know not to get scammed? Lucas: You don't verify them. You verify the platform's ability to filter them. The risk shifts from choosing the right vendor to trusting the gatekeeper. Think about Amazon. You rarely read individual reviews anymore for basic goods. You just check if it has the 'Amazon's Choice' tag or comes with Prime shipping. That is a trust proxy. Luna: But doesn't that concentrate too much power in the platform? If Amazon decides to demote a seller, they have no recourse. Lucas: That is the exact tension. In a transparent rating world, sellers had some leverage. They could campaign for better reviews or explain context. In an opaque world, the algorithm is the boss, and the boss doesn't explain its decisions. We are seeing this play out in real time with gig economy workers who get deactivated without knowing which specific metric failed. Luna: It feels like we are trading autonomy for convenience. Lucas: Convenience is the currency here. Most users will happily give up the right to judge if it means they don't have to think about judging. The platform takes on the cognitive load of trust. But there is a dark side to this efficiency. Luna: Which is? Lucas: Homogenization. When everyone optimizes for the same hidden algorithm, services start to look identical. A hotel chain, a boutique B&B, and a vacation rental all start offering the same standardized amenities because that is what the trust model rewards. You lose the weird, risky, high-variance options that usually offer the best value. Luna: So we get safer, but also drier experiences. Lucas: Dry is a good word. Let me give you a concrete example from the creative services sector. Upwork tried moving toward a project-based scoring system where freelancers aren't rated globally, but only on specific past projects within your category. It sounded smart. It reduced bias. Luna: Did it work? Lucas: It actually killed engagement for new entrants. Without a global reputation score, a new freelancer has zero social proof. The old system let you build a name over time. The new system requires you to be perfect on day one to get any visibility. So the platform ended up protecting incumbents even harder. Luna: That sounds like a classic network effect trap. Once you're in, you stay in, regardless of whether you're actually the best fit. Lucas: It is exactly that. And it creates a false sense of security for the buyer. You click 'hire' because the badge says 'Top Rated,' but you have no idea if that rating reflects skill or just longevity. Longevity is not a proxy for competence in fast-moving fields like coding or digital marketing. Luna: So the badge becomes a status symbol rather than a quality indicator. Lucas: Status. Yes. And platforms love status because it is sticky. People hoard badges. They optimize their profiles to collect them. It gamifies the supply side in a way that benefits the platform's retention metrics, not necessarily the consumer's outcome. Luna: I guess the question is whether consumers are smart enough to push back against this opacity. Lucas: Most aren't. But a segment is starting to migrate to direct relationships. We are seeing a rise in niche communities where trust is built through community verification rather than platform algorithms. Discord servers, specialized Slack groups, private newsletters. These are walled gardens of trust. Luna: Because the cost of exclusion is higher, the quality of interaction stays high. Lucas: Exactly. If you get kicked out, you lose access to everything. On Uber, if you get de-rated, you just use Lyft. The portability of reputation across opaque platforms is near zero. That locks you in. Luna: It makes switching costs effectively infinite. Lucas: Infinite for the average user. For enterprises, it is manageable. Large companies have procurement teams that bypass these badges entirely. They negotiate contracts directly. The opacity of the consumer marketplace is really a feature for small-scale, low-stakes transactions where due diligence is too expensive. Luna: So we are left with a two-tier system. One for the masses who trust the badge, and one for the pros who ignore it. Lucas: Two tiers. And the platform profits from both. They take a cut from the badge-buyers for the convenience, and they take a cut from the pros for the liquidity they provide. It is a perfect extraction model. Luna: It is efficient, but it feels brittle. What happens when the algorithm makes a mistake? Lucas: That is the crisis point. Because there is no public score to contradict the platform's decision, the user has no narrative. They cannot argue 'but my previous ten guests loved me.' They can only argue 'the machine thinks I am bad.' And machines do not negotiate. Luna: That must be terrifying for someone whose livelihood depends on that single algorithmic decision. Lucas: It is. And it is why we are seeing more regulatory scrutiny on 'algorithmic management.' The EU is already drafting rules that require platforms to provide some level of transparency into how these trust scores are calculated. Not the full code, but the key variables. Luna: So forcing the black box to crack open just a little. Lucas: Just a little. Enough to prove it isn't random. But I doubt they will force full transparency. Platforms will resist that fiercely because transparency invites competition. If I can see exactly how you are scored, I can beat your score. Luna: And then your moat disappears. Lucas: The moat is the mystery. As long as you don't know how the sausage is made, you assume it is safe. Once you see the ingredients, you might decide you don't want to eat it. Luna: A powerful metaphor. And a chilling reality for anyone working on these platforms. Lucas: It changes the nature of work itself. You are no longer competing with other humans. You are competing with a ghost. A set of weights and biases that you cannot see, touch, or influence. Luna: Do you think this trend will reverse? Will users demand their stars back? Lucas: Unlikely in the short term. Convenience always wins until something goes wrong. And when it does go wrong, the backlash is loud but temporary. Then life goes on, and the badge still gets you the booking. Luna: So we keep trading pieces of our agency for the ease of clicking 'book now.' Lucas: Piece by piece. Until one day we realize we forgot how to judge for ourselves. Luna: That is a sobering thought. It makes me wonder if the next wave of innovation will be in rebuilding human to human trust outside these systems. Lucas: Probably. Direct relationships are making a comeback, but they are expensive. Building a brand that doesn't need a badge is hard work. Luna: Harder than just optimizing for the algorithm. Lucas: Much harder. But maybe that is the only way to keep some skin in the game. Luna: Speaking of skin in the game, if these conversations are useful for what you are building or running, consider supporting the show directly. Lucas: Listener support keeps this network ad-free and allows us to dig into these structural shifts without chasing clicks. You can find us at buy me a coffee dot com slash fexingo — that's F. E. X. I. N. G. O. Luna: It helps us keep the lights on and the research deep. Thanks for keeping us going. Lucas: Back to the trust model. The real danger isn't just that we forget how to judge. It's that we start believing the platform's version of reality is the only objective truth. Luna: When the map replaces the territory, navigation becomes impossible. Lucas: Exactly. And we are all just wandering around in the dark, following the glow of a badge we don't understand. Luna: Maybe the first step is turning off the glow and looking at the person behind the screen. Lucas: Or at least asking what algorithm put them there in the first place.