Latest / Marketplace Businesses with Fexingo: Two-Sided Networks, Liquidity, and Take Rates / Why Marketplaces Raise Take Rates When They Offer On-Time Guarantees
Transcript
- Lucas: If you've ordered delivery through DoorDash recently, you might've noticed a new line in the checkout flow — a promise that says 'your order will arrive by this time, or you get credits.' Luna: Right, they rolled that out more aggressively last year. It's almost like a mini-insurance policy baked into the transaction. Lucas: Exactly. And what's interesting is that this on-time guarantee isn't just a customer experience play — it's a take-rate lever. When a marketplace offers a concrete reliability promise, it can charge a higher commission without losing users, because the guarantee changes the value proposition. Luna: So you're saying the guarantee itself justifies a higher fee? That feels counterintuitive — usually adding a feature costs money, not makes it. Lucas: It does cost money — DoorDash has to subsidize credits when orders are late. But the key is that the guarantee shifts user psychology from price sensitivity to reliability value. If I know my lunch will arrive by 12:15 or I get $5 off my next order, I'm less likely to comparison-shop on delivery fees. That gives the platform pricing power. Lucas: Quick honest thing — this episode, like all our episodes, is made possible by listeners who chip in through buy me a coffee dot com slash fexingo. No ads, no sponsors, just a small group keeping the show going. If these conversations are useful for what you're building or running, that link is the whole engine. Luna: Yeah, it's a simple model — listeners support the show, we keep it ad-free and focused on the actual mechanics. Appreciate everyone who's part of that. Lucas: Alright, back to guarantees. DoorDash's 'DashPass promise' — which launched nationwide in early 2025 — guarantees delivery within a stated window for DashPass members. Non-members see a similar guarantee on select orders. The result? DoorDash raised its take rate on guaranteed orders by roughly 2 to 3 percentage points compared to non-guaranteed ones. Luna: That's a significant bump. But how do they keep the guarantee from eating into margins? If every order is late, they'd be handing out credits all day. Lucas: They manage it through a combination of dynamic pricing and order bundling. During peak hours, the guarantee window widens — say, 30 minutes instead of 20 — which gives the system more flexibility. And they batch orders going in the same direction, so a single driver handles multiple deliveries, reducing per-order cost. Luna: So the guarantee is tighter when demand is low and looser when it's high. That's basically surge pricing in reverse. Lucas: Exactly. And that creates a data moat — the more orders they fulfill with guarantees, the better they get at predicting delivery times. Instacart does something similar with its 'Priority Delivery' tier. For an extra fee — about two to three dollars — customers get a narrower delivery window and a guarantee that a shopper will start picking their order within 30 minutes. Luna: Instacart's take rate on those priority orders is about 20 percent, compared to 15 percent on standard orders. That five-point spread is basically pure margin because the cost of the guarantee — the occasional late order — is small relative to the extra revenue. Lucas: Right. The guarantee becomes a signaling mechanism. Customers who choose it self-select as less price-sensitive, so the platform can charge them more. And the data from those orders — which items take longest to pick, which routes cause delays — improves the algorithm for everyone. Luna: There's a risk, though. If you over-promise on delivery times, you end up paying out credits frequently, which erodes the take-rate gain. DoorDash learned that the hard way during early tests in 2023 when they guaranteed 20-minute delivery in dense urban areas and saw late rates above 15 percent. Lucas: That's why they now set the guarantee based on historical data at the restaurant level — not a blanket promise. A McDonald's in midtown Manhattan might have a 25-minute guarantee, while a sushi place in a suburban strip mall gets 40 minutes. The take rate adjusts accordingly, higher for tighter windows. Luna: So the guarantee itself becomes a product tier. You're effectively selling reliability as a premium feature, and the take rate is the price. Lucas: Exactly. And this isn't just for food delivery. Think about marketplaces like Thumbtack — they introduced a 'Job Guarantee' where if a service provider doesn't show up or does poor work, Thumbtack refunds up to $1,000. That guarantee lets them charge a higher commission — around 20 percent instead of the typical 15 percent for similar platforms without guarantees. Luna: That's a 33 percent increase in take rate just for adding a guarantee. And the actual claims rate is low — maybe 2 percent of jobs — because the guarantee itself incentivizes providers to show up and do quality work. Lucas: Right. It's a virtuous cycle. The guarantee builds trust, which attracts more users, which gives the platform more data, which lets them price the guarantee accurately, which lets them raise the take rate without losing volume. The key is that the guarantee must be credible — if customers don't believe you'll honor it, it doesn't work. Luna: And credibility requires a financial buffer. DoorDash sets aside a reserve fund for guarantee payouts, typically around 0.5 percent of gross transaction value. That's small enough that the take-rate uplift — say 2 percent — more than covers it. Lucas: So you're spending half a percent to gain two percent. That's a 4x return on the guarantee investment. And because the guarantee improves the customer experience, repeat purchase rates go up, which further lifts lifetime value. Luna: I wonder if we'll see more marketplaces adopt this model — especially in categories like professional services or home services where trust is a bigger barrier than price. Lucas: I think we already are. TaskRabbit has a 'Tasker Happiness Pledge' that refunds if you're not satisfied. They charge a service fee of around 15 percent, compared to 10 percent for similar platforms without guarantees. And that's a direct take-rate lift. Luna: So the pattern is clear: guarantee equals trust equals pricing power. But there's a ceiling — if you raise the take rate too high, even with a guarantee, users will churn. What's the limit? Lucas: Good question. Research from marketplace platform experts suggests the ceiling is around 25 to 30 percent take rate for most two-sided marketplaces, even with strong guarantees. Beyond that, the cost to the customer outweighs the value of the guarantee. But within that range, a well-designed guarantee can add 3 to 5 percentage points of take rate. Luna: So the next time I see a guarantee on a delivery app, I'll know that two-dollar fee isn't just for the guarantee — it's the platform's way of capturing a bit more value from my trust. Lucas: Exactly. And next episode, we'll look at how marketplaces use refunds and return policies to build trust — and how that affects take rate in a different way.