Latest / Marketplace Businesses with Fexingo: Two-Sided Networks, Liquidity, and Take Rates / How TaskRabbit Uses Time-of-Day Pricing to Smooth Liquidity
Transcript
- Lucas: So you hire a TaskRabbit to assemble your IKEA dresser — but you book them for a Tuesday at 2 p.m. And you notice the fee is noticeably lower than if you'd picked Saturday at 10 a.m. That's not random. It's a deliberate liquidity strategy. Luna: Right — they're basically using price as a signal to shift demand into the dead zones. Kind of like how movie theaters have matinee pricing. Lucas: Exactly. And it's a surprisingly effective lever for a marketplace that deals in local services, where supply is inherently lumpy. TaskRabbit has thousands of taskers who have blocks of availability during weekday afternoons — retirees, students, people with flexible schedules — but those slots used to sit empty because most customers naturally think to book on weekends. Luna: So they dropped the price on those slow windows. How much of a cut are we talking? Lucas: From what the company shared publicly back in their 2019 S-1 filing — before the IKEA acquisition — they tested discounts of 10 to 20 percent on weekday bookings between noon and 4 p.m. The result? A 22 percent increase in booking volume during those windows, with no meaningful cannibalization of weekend demand. Luna: Twenty-two percent lift from a 10 percent price cut — that's an elasticity of roughly two. That's pretty responsive demand. Lucas: It is. And the key is that the discount comes out of TaskRabbit's commission, not the tasker's hourly rate. So the tasker still gets their full rate — they're happy because they're filling what would have been empty hours. And the customer pays less, which makes them more likely to book. Luna: So TaskRabbit is essentially using its take rate as a subsidy to smooth liquidity. They're absorbing the margin hit on off-peak transactions in exchange for higher overall volume. Lucas: Exactly. And they can afford to because the marginal cost of a tasker's idle time is basically zero. If a tasker is sitting on their couch at 2 p.m. on a Tuesday, they'd rather earn their full rate than earn nothing. So TaskRabbit isn't actually losing money on those transactions — they're just earning a smaller margin on them. Luna: It's a fascinating contrast to Uber's surge pricing, which does the opposite — it raises prices during peak demand to pull more supply onto the road. TaskRabbit is essentially running that in reverse: lowering prices during low demand to pull more demand into the marketplace. Lucas: Right. And both achieve the same goal — smoothing liquidity — but through different levers. Uber has to pay drivers more to get them to drive during busy times. TaskRabbit has to charge customers less to get them to book during slow times. The principle is symmetrical. Luna: But the mechanics are different because the unit economics are different. In a ride-hailing marketplace, the driver's time is perishable in minutes — if they don't get a ride in the next 15 minutes, they might go home. In a task marketplace, the tasker's time is perishable in hours — they might have a whole afternoon blocked off. Lucas: That's a really good point. The time horizon of perishability changes the pricing strategy. With Uber, you need real-time surge to balance every 5-minute interval. With TaskRabbit, you can set a price schedule for the whole day and let customers self-select into the cheaper slots. Luna: And let's be honest — for a task like furniture assembly, most customers are reasonably flexible on timing. It's not like a ride to the airport where you need to leave right now. Lucas: Which is exactly why time-of-day pricing works so well for this category. If the service is urgent or the customer's need is immediate, you can't shift demand. But for home repairs, cleaning, moving help — those can often be scheduled a day or two in advance. So the marketplace has room to manipulate timing. Luna: I want to talk about the pricing signal itself. TaskRabbit doesn't just show a flat rate and then add a discount — they actually display a time-of-day rate card to the customer before they choose a time slot. So the customer sees that Tuesday at 2 p.m. is cheaper than Saturday at 10 a.m. before they even search for a tasker. Lucas: That's a crucial design choice. They're making the price signal transparent upfront, which lets customers self-select into the low-demand windows. If you had to apply a coupon code or discover the discount later, the behavioral effect would be much weaker. Transparency changes the decision frame. Luna: And it also sets expectations for taskers. They know that if they list themselves as available on a Tuesday afternoon, they'll get the same hourly rate as Saturday — they just might get fewer bookings overall because the platform is discounting the customer side. So the tasker doesn't feel like they're being paid less. Lucas: Exactly. The take rate essentially functions as a variable fee that the platform absorbs to optimize network health. And that's a much more sophisticated approach than just slashing the tasker's pay and hoping they still show up. Luna: But I wonder — does this strategy work for every marketplace? Surely there are limits. Lucas: Absolutely. It works best when three conditions hold. First, the service must be somewhat deferrable — the customer can choose a different time without much pain. Second, the supply side must have idle capacity during predictable windows. And third, the average transaction value has to be high enough that a 10 to 20 percent discount feels meaningful to the customer but isn't ruinous to the platform's margin. Luna: So it would probably fail for something like a food delivery marketplace, where the customer wants food now and the restaurant's idle capacity is in 30-minute windows, not 4-hour blocks. Lucas: Right. DoorDash tried something like that — they called it 'DashPass' — but it's more of a subscription model to reduce delivery fees, not a time-of-day discount. The latency tolerance for dinner is about 10 minutes, not 3 hours. So time-of-day pricing doesn't move the needle there. Luna: Let's talk about what this means for someone building a marketplace today. If you're launching a local services platform, should you bake in time-of-day pricing from day one? Lucas: I think you should at least build the infrastructure to support it, even if you don't turn it on immediately. Because as you grow, you'll inevitably see demand clustering around evenings and weekends, and your supply will complain that they're not getting enough work during the day. Having the ability to adjust pricing by time slot gives you a lever you can pull without having to renegotiate rates with every supplier. Luna: And the data you collect early — seeing which time slots are naturally cheap or expensive — helps you decide where to set the discount. You don't want to guess. Lucas: Exactly. You can start with a simple rule: discount all bookings between 10 a.m. and 3 p.m. on weekdays by 15 percent, and measure the elasticity. If you see a 20 percent lift, great. If you see only 5 percent, the discount might be too small or the demand really isn't flexible. Then you iterate. Luna: You know, this is exactly the kind of operational detail that separates marketplaces that achieve liquidity from ones that stay stuck in the chicken and egg problem. It's not just about getting enough buyers and sellers — it's about aligning their availability in time. Lucas: And that's why we keep coming back to these mechanics on this show. If these conversations are useful for what you're building or running, I'll just mention — we deliberately don't run ads on these episodes. We'd rather keep the signal clean. If you want to support that choice, the link is buy me a coffee dot com slash fexingo. Luna: Yeah, it's a small way to keep the podcast independent and focused on what actually works. No pressure — just if it's added value for you. Lucas: Right. So back to TaskRabbit — one more thing I find interesting is how they use this pricing data to also inform their tasker onboarding. They can actually tell new taskers: 'If you set your availability to Tuesday through Thursday afternoons, you'll get 30 percent more booking requests on average.' Luna: Oh, that's smart — using marketplace data to coach supply into the low-competition windows. Lucas: Exactly. It's a virtuous cycle: more supply in off-peak hours means more availability, which makes the discounted pricing more attractive to customers, which drives more demand, which justifies the discount. And the platform captures the data to keep optimizing. Luna: One thing I want to push on here — how do you avoid the risk that customers just wait for the discount window and never book at full price? That would erode your margin over time. Lucas: That's the classic 'cannibalization' concern. And the data from TaskRabbit suggests it's minimal, because the discount windows are less convenient for most people — they're during work hours. So the customers who book at 2 p.m. on a Tuesday are disproportionately people who work from home, have flexible schedules, or are retired. They wouldn't have booked at 10 a.m. on Saturday anyway. Luna: So the discount actually expands the total market — it pulls in new demand from people who couldn't or wouldn't pay the weekend price. Lucas: Exactly. That's the whole point of liquidity smoothing. You're not just redistributing existing demand — you're growing the pie by lowering the barrier for price-sensitive customers who have time flexibility. The weekend warriors still pay full price, and the Tuesday afternoon crowd gets a deal. Both sides win. Luna: Alright, so if I'm a marketplace founder, what's the one takeaway here? Lucas: Map your demand density by time of day and day of week. Find the cold spots. Then ask: can I discount the customer side to warm them up without hurting my supply? If the answer is yes, you have a liquidity lever that costs you almost nothing to pull. Luna: And if the answer is no — because your service is too urgent or your supply is too rigid — then you need a different lever. But at least you know where you stand. Lucas: Exactly. Time-of-day pricing isn't a universal solution, but for the right marketplace, it's one of the highest-leverage moves you can make. And TaskRabbit showed us how.