Hey folks, welcome back to Paradox Weekly!

Last Saturday, gig drivers across Delhi, Mumbai, Bengaluru and Hyderabad went offline on Ola, Uber and Rapido for six hours.

They called it the "All India Breakdown."

Two days before the strike, the government formally launched Bharat Taxi, a cooperative ride-hailing model backed by Amul, IFFCO, NABARD and five other national cooperative institutions.

With zero commission and no surge pricing; 3 lakh drivers and roughly 10 lakh users have already registered, with 2.91 lakh rides completed so far.

That tells you where we are.

Before we go further, quick poll:

Getting access to some of the media as the episode is about to air!

Yup, day or night our students are at it : )

You’ve probably seen the video by Pooja Chhabda on Zepto.

She checked the price of 500 grams of grapes: ₹65 on her Android, ₹146 on her iPhone.

Zepto hasn't said a word publicly.

It gets worse…

MediaNama tracked prices across Blinkit, Instamart and Zepto from December 2024 to January 2025. iOS was consistently higher.

Infrequent users paid more than daily users.

People confuse this with surge pricing. Surge is when prices go up because demand is high.

What these platforms actually run is something called first-degree price discrimination: every user sees a price calculated specifically for them, based on their device model, order frequency, cart abandonment rate, pin code, and time-of-day patterns.

If you order daily, the platform shows you lower base prices to keep the habit. If you order once a week, you see higher prices because the platform gets one shot at margin.

If you never abandon a cart, the algorithm already knows you'll pay whatever it shows you.

A Columbia Business School study analysed 31,000 Uber trips from a single driver between 2018 and February 2025.

Uber's take rate climbed from about 32% when "upfront pricing" launched in 2022 to upwards of 42% by end of 2024.

On a third of those trips, Uber took home more than the driver did.

An Oxford study of 1.5 million UK trips found the same pattern: take rate up, average driver pay adjusted for inflation down from £22.20 to £19.06 per hour.

What’s funny is that the app shows the rider and the driver different numbers.

The rider thinks the fare is high because of demand. The driver thinks the payout is low because of supply.

Neither side sees what the other is getting.

Before Ola and Uber, you haggled with an auto driver who refused the meter.

He quoted ₹500 for a ₹200 ride. That was unfair.

But you knew three things: the price, who set it, and that you could walk away.

The app was supposed to fix that. Price on screen, no negotiation.

Except now you don't know how the price was calculated, what someone else is paying for the exact same ride, or what the driver is earning from your fare.

People would literally pay extra to stop feeling gamed.

Here's the paradox: we went to these platforms because they promised the real price.

The screen was supposed to be the solution.

Instead it became the most effective pricing weapon ever built: because it looks like transparency while operating as the exact opposite.

CCI fined MakeMyTrip-Goibibo ₹223 crore and OYO ₹169 crore for price parity clauses: hotels were contractually prevented from offering lower prices on their own websites.

AND OYO changes room prices 46 million times a day.

Swiggy and Zomato charge restaurants 25-35% commission, so restaurants inflate app prices to cover it, then the platform stacks delivery fee, platform fee and GST on top.

BookMyShow had 150,000 Coldplay tickets, 13 million people tried to buy them, and those same tickets showed up on Viagogo at ₹7.7 lakh. Police investigated and closed the case, but a company holding ~75% of online entertainment ticketing watched its own inventory flood the secondary market at 30x face value.

Every sector seems to have the same play: show a price to build the habit, build the habit to control demand, control demand to squeeze supply.

Namma Yatri charges ₹25/day or ₹9 per ride and the driver keeps the full fare. 100 million rides, 600,000+ drivers, ₹1,600 crore in driver earnings all on $11 million in total funding.

Ola and Uber burned billions to reach comparable numbers.

But here's my problem with the "Namma Yatri will save us" narrative: annual revenue is ₹23.5 crore.

They're operationally profitable in only two states.
Ride conversion drops from 25% in Bengaluru to 7% everywhere else.
Bharat Taxi does about 10,000 rides a day while Rapido does 4.3 million.

Bharat Taxi's total completed rides to date is less than what Rapido does before lunch on a Tuesday.

And Uber, Ola and Rapido have already started copying the subscription model in response.

We went to these platforms to kill information asymmetry.

The auto driver knew more than you about what the ride should cost, and the app was supposed to equalize that.

Instead, the platform now knows more about both sides than either side knows about the transaction. And it profits from keeping both in the dark.

We replaced a problem we could see with one we can't.

Hit reply: what’s the craziest price difference you’ve experienced across websites / devices?

I read every email.

Until next week,
Pratham