Case study on ride-hailing apps: improvements for drivers and passengers

I have been an avid user of uber, namma yatri and rapido for my adventures within the city. Quick rides to get around with the family and friends for get-aways and what not, but there are two instances which I felt could be improved from all these services.

---

Observed issue 1: Notification of next trips in driver's app

Cab drivers when nearing a completion of the trip, are prompted with the next trip details in a hazardous manner, here's why

• A constant buzzing sound.

• Chunks of text which contain full length pickup and drop addresses in smaller text and the pay in bigger text, taking up half the space.

• The above two forces the driver to pay a lot more attention on the phone to understand where the pickup and drop locations are so they have to put in some focus on reading all the while driving which in itself takes up a lot of cognitive load.

Now throw in other ride hailing apps as well which drivers would have signed up for more trips, combine 3-4 apps constantly pinging on a low budget phone which generally have lesser processing powers. This can be an annoyance for everyone and potentially crash the phone of the driver.

Here's a screen shot of the next trip ui screen of uber which I could find from this youtube video. You can also get to listen to the sound notifying the driver about the trip in the video.

Proposed solutions

Quick fix:
Instead of annoying the user into taking an action on the trip, use a voice output to notify the driver about the trip with just the key details, example:

"New trip available, pickup is 1km away from drop location which will pay 138 rupees and 39 paisa"

The drop location and other details can be expanded on if the user is interested in, I believe by now they would also just intuitively know how far a trip is just by the price as well.

A long term solution:
This is where AI voice assistants can shine I believe. A true assistant which can help drivers with the required info via conversations in the driver's preferred language.

A hypothetical conversation flow between a driver (Raj) and uber's ai assistant (let's say Rueb? I just rotated the word around in the app's name):

Rueb: "Hey Raj, next trip is available which is 4 mins away from our current destination and pays 138 rupees and 39 paisa, to know more just say more info and to accept the ride say accept ride"

Raj: "more info"
[considering language barriers and other situations, it's probably best if we ask the drivers to say "Rueb more info"]

Rueb: "A 138 rupee trip which has a 5.6kms of expected driving distance, the drop location is near mango, jamshedpur which can last about 18 mins of total driving time. View the app for more accurate details of the location"

The key take away from this conversation

• First we let them know about the distance to go to the new pickup location along with the fare

• If they are interested then we give out key details of the drop location like driving distance, minimized address so that the driver get's a general idea of the distance and not be overloaded with info from long addresses. All the while again reminded of the fare which is a key motivator.

• In the second response from the assistant there is no mention of the pickup location because we have already mentioned it in the first step, repeating the address again could be unnecessary.

The main attempt here is that the first few conversations should help drivers resolve quickly with minimal sentences.

Metrics for success

• Faster driver search for riders, the whole AI conversation feature should help drivers decide if they would like to take up the ride, thus resulting is faster search for users.

• Lesser cognitive load / distraction for drivers which is crucial for safety.

• Lesser reliance on UI based activities, can cut out the need for textual messages and calling features, just say "Rueb notify the rider that I'm at the location" and the rider get's the notification. Making it all hands free and easy for both riders and drivers.

---

Observed issue 2: Subscribed rides

This is rather a simple one, moms with kids who are signed up for extra-extra-curricular activities need a hand. Every time booking an auto to the same old destination like a cricket coaching class or a pickup and drop on a weekend only singing class could all be made scheduled / subscribed.

• Picking up the phone 25 mins prior to the time they actually leave for the class which is usually 30 mins before, so that they have some luck in finding an auto / cab just in time to reach the destination on time, is not happening / cannot happen every day.

• Uber's reserve works only for a specific one time instance, not a recurring ride for a set period of time.

Proposed solution

I whipped out a quick mock for a subscribed ride ui screen on v0 to just keep things simple, here's the link to the chat

https://v0.dev/chat/cab-ride-scheduler-2eqzLlk3zMv

Here's a screenshot for a quick analysis

• let the user pick which days the rides are needed and time

• Recurrence is crucial as this is not always guaranteed to be used for weeks to come

• location fields for the pickup and drop

All of this does come with a lot of edge cases as well, like not enough drivers availability for certain locations on certain days, On the dot cancellation for the day from the rider due to change in plans thus causing a loss for assigned drivers.

Some ways to reduce the damages by these edge cases:

• Penalty for not cancelling the ride before an hour for the pickup time.

• Assign a driver within 4-5 hours before the pickup time, else notify the user through a call and offer other available solutions if possible. This should give enough time for the user / rider to consider their options and plan the day accordingly.

• Failure from driver to be at the pickup spot, have a strike system for this and offer an alternative available ride service for the user at the same cost. Customer satisfaction at such rare cases would help with retention and building loyalty.

Metrics for success

• Increased revenue as this pushes the user to use our services more.

• Increased rides per day for drivers so that they have more stable earnings throughout the day which helps with retention.

• Potential to expand avenues in schools and teaching centers for collaboration to provide stable transportation for their customers.

---

By addressing these issues, I believe ride hailing apps can improve the collective experience of both drivers and riders. I hope AI based voice assistants and subscribed rides are something we see in the future and help generate revenue for all the parties involved, all the while helping drivers be less exhausted with the daily stress they take up in this concrete jungle.

Credits:

• Cover photo: Photo by Tima Miroshnichenko: https://www.pexels.com/photo/person-using-black-android-smartphone-6169870/

• Originally published on peerlist.