Latest / Pricing Power Podcast with Fexingo: How Businesses Set Prices, Raise Margins, and Win Customers / How Uber Uses Surge Pricing to Balance Supply and Demand
Transcript
- Lucas: Let's start with a number: in December 2014, a 16-hour hostage siege unfolded at a café in Sydney's Martin Place. Lucas: Police cordoned off the area, public transport was disrupted, and thousands of people needed to get home. And Uber — which was still relatively new in Australia — activated surge pricing. Fares spiked to four times the normal rate. Luna: I remember the backlash. People called it price gouging during a tragedy. The hashtag #UberSurge trended, and the company took a massive PR hit. Lucas: Right. Uber's defence was that higher prices attract more drivers into the area, which means more rides get completed. If prices stay flat during a sudden demand spike, you get a shortage — long wait times or no rides at all. Luna: But that's a hard sell when someone's trying to flee an emergency. It feels like the algorithm doesn't have a conscience. Lucas: And that tension — between economic efficiency and human empathy — is the core of surge pricing. So today I want to dig into how Uber's system actually works under the hood, and whether the model has evolved. Lucas: The basic logic is straightforward. Uber uses a real-time heat map that shows where rider requests are concentrated and where available drivers are. When requests in a zone outnumber available drivers, the algorithm raises the multiplier. Luna: And the multiplier is applied to the base fare, right? So it's not just a flat dollar increase — it scales the entire trip cost. Lucas: Exactly. The base fare covers time and distance, plus a booking fee. Surge multiplies that total by something like 1.2x to 4x, sometimes higher in extreme cases. The idea is to nudge drivers — 'hey, head to this neighbourhood, you'll earn double per mile.' Luna: And it works. Studies show that a 1.1x surge increases driver supply by about 5 to 10 percent in a given area within minutes. Lucas: But here's the thing — the multiplier isn't just about the current moment. Uber uses predictive algorithms that forecast demand 30 to 60 minutes ahead, based on historical data, weather, events, even time of day. Lucas: So if a concert lets out at 11 PM, the system might start gradually increasing prices before the crowd even opens the app. That preemptive surge smooths the spike. Luna: That makes sense from a logistical standpoint. But for the rider, seeing a 2.5x fare on a normally $15 trip feels like a gut punch. And there's no negotiation — it's take it or wait. Lucas: Right, and that's where the controversy lives. Uber's early approach was pure algorithm — no human override. After Sydney, they introduced 'price caps' during major emergencies, but those are manually triggered and not always fast enough. Lucas: They also launched a feature called 'Upfront Pricing' in 2016, which shows the exact fare before you confirm. That reduced sticker shock because the multiplier is baked in — you see a total, not a percentage. Luna: Interesting. So the presentation changes the perception, even if the underlying math is the same. Behavioral pricing. Lucas: Exactly. But I want to push back on something. Critics often frame surge pricing as pure profit extraction. In reality, Uber's take rate is roughly 20 to 25 percent of each fare. The rest goes to the driver. So a surge fare means the driver gets paid more, not just Uber. Luna: That's a fair point. But it also means that during a surge, Uber's absolute revenue per trip increases. So they do benefit from the higher volume of pricier trips. Lucas: True. But the key insight is that surge pricing is a supply-side incentive, not just a demand-side tax. Without it, drivers wouldn't relocate to crowded areas because their per-mile earnings would drop. The system balances the market. Lucas: And there's a second layer: Uber also uses 'surge shroud' in some markets — they show riders a flat dollar 'boost' instead of a multiplier, like '$5 extra' rather than 1.5x. Same economic effect, different framing. Luna: So they're A/B testing just the language of pricing. That's pretty sophisticated. Lucas: It is. And it highlights a broader shift. Competitors like Lyft initially capped surge at 2x, then moved to a more dynamic model. But both platforms now blur the line between surge and 'primetime' or 'event pricing'. Lucas: Let's talk about one more concrete example: New Year's Eve. In 2024, Uber fares in major US cities hit 4.2x around midnight. But average wait times stayed under 8 minutes, compared to 15+ minutes on a normal Friday night. Luna: Wow. So the surge actually delivered on its promise — more rides, faster. Lucas: Exactly. And drivers earned about $35 to $45 per hour during that period, versus $20 normally. So the system worked. But it only works if enough riders are willing to pay the premium. And some aren't — they choose to walk, take public transit, or wait. Luna: And those people effectively self-select, reducing demand pressure. That's the elegant part — the price becomes a rationing mechanism. Lucas: Right. So the question becomes: is surge pricing fair? I think the answer depends on transparency and context. In a normal Saturday night, it's just supply and demand. In a crisis, it feels predatory. Luna: And that's why Uber now has a 'peace of mind' policy — they automatically cap surge during declared emergencies and refund any rider who was charged more than normal during a disaster. Lucas: That's a good evolution. But it also shows that pricing algorithms need a human ethical framework, not just an optimisation function. Luna: Speaking of evolution — before we wrap, I want to shout out something. These conversations we have every week take real time to research and produce. A couple of dollars a month from listeners genuinely keeps this show ad-free and independent. Lucas: Yeah, if these episodes have ever given you a new way to think about a business problem or helped you make a smarter decision, buy me a coffee dot com slash fexingo is the best way to support that. No pressure, but every bit helps. Luna: And we mean it — it's because of that support that we don't run ads or chase clicks. Just solid content. Alright, back to pricing. Lucas: So where is surge pricing headed? Uber is experimenting with 'subscription' models that include surge protection — pay a monthly fee and get locked-in base fares regardless of conditions. Luna: That's fascinating. It flips the model from per-trip dynamic pricing to a fixed fee for certainty. Almost like an insurance premium. Lucas: Exactly. And if that catches on, it could reshape the whole ride-hailing pricing structure. But for now, surge remains the engine that keeps the market liquid. Luna: I think the big takeaway is that surge pricing isn't good or bad by itself — it's a tool. The ethics depend on how and when you use it. Lucas: Well said. Next episode, we'll look at how a completely different industry handles dynamic pricing — stadium concessions. You'll never look at a $12 hot dog the same way. Luna: Looking forward to it.