How Does Uber Match Riders With Drivers? Who Gets Assigned To Who?

A rider requests a ride, and the nearest driver is pinged every time, right? Well that’s the way it used to be, but not any more. Joe here at The Rideshare Guy, and today we’re going to discuss how Uber matches riders with drivers.

The Uber matching system used to be a straight forward

When Uber first launched, their matching system was straight forward. A rider would make a request, and the nearest driver would be pinged with that request. But at some point that changed. Uber didn’t make an official announcement when this happened, but they recently published a article on their website showing how it works.

Matching with the nearest driver sometimes isn’t the most efficient way to match riders with drivers. So Uber thought of a way to make this a more efficient process by introducing what they call batched matching.

The new matching system: Batched matching

Let’s take a look at an example, and see how it would play out under the old system and under the batched matching system.

Under the old system rider A requests a ride, and the nearest driver is two minutes away. Then driver A accepts the request, and a match has been made. Nearby, though, rider B requests a ride, and the nearest driver to them is now nine minutes away. So total wait time for those two requests is 11 minutes.

Now if we take a look, driver A was actually only four minutes away from rider B, and driver B was actually four minutes away from rider A. Using batched matching, after a request has been made, the system will think a little bit instead of immediately matching a rider with the nearest driver. The goal is to reduce wait time as much as possible.

The current ride matching system aims to reduce overall wait times

As in the example, instead of matching rider A with the nearest driver, the system thinks a bit, and instead matches more efficiently by understanding that it can reduce the total wait time by three minutes if it matches both requesters with the opposite driver. So rider A will have to wait two more minutes for driver B, but driver B will reduce their ETA by five minutes. And rider B’s wait time will be reduced by five minute, but driver A’s ETA will increase by two minutes.

Rider A and driver A are effected somewhat negatively by batched matching for this particular situation, but rider B and driver B are effected positively, and overall the more this plays out over many ride requests, it is a net positive for everyone. And Uber states that every day batched matching is saving riders and drivers 10 years of time compared to the old method of matching. That’s pretty incredible.

We’re not sure when this change to batched matching happened, but for those of you that drive for Uber, have you noticed a difference? I’m primarily a Lyft driver. I’m hoping that Lyft either has implemented or will implement a similar system, because I still get many inefficient requests on Lyft.

What does the matching system mean for us drivers?

A more efficient system is just better for everyone involved. I’ve been a driver for nearly five years now, and the Uber system has been made much more efficient since I began, even within the last year. From no-wait trip requests to re-routing and now batched matching, Uber is constantly trying to make their system more efficient.

And they’re going to continue to do this, because the more efficient this system is, the happier riders, happier drivers, less ETAs, lower wait times, and they’re going to continue to do this, because if Lyft doesn’t make the same changes, it gives Uber a competitive advantage. Ultimately, you want the lowest possible ETAs and wait times, and Uber has a group of data scientists that are working towards making the system even more efficient.

Thanks again for watching and reading. Please like, comment, or subscribe, and drive safe!