From Confusion to Clarity

Solving Fare Shock with Time-Based Pricing

Fare shock isn’t about the money. It’s about the moment.

We knew we had to explain it better, not just calculate it smarter.

We wanted to do things differently.
What if we made fare updates feel fair, not frustrating?

Other apps charged for time too — they just didn’t talk about it.

They add it. You notice it. Too late.

— Traffic charges,” the app told him, right after the ride ended.

— He sighed. “It’s not about the money. It’s just... I felt tricked.”

— “Why did the fare increase?”

— “But no one told me that before. Wasn’t this supposed to be ₹80?”

The fare made sense. The moment didn’t.

Let's start with a story

We spoke to a frequent cab user in Namma Bengaluru.
His ride was just 4 km — the kind he’d taken dozens of times.
But this time, it took 40 minutes instead of 15.
The fare? ₹112 instead of the usual ₹80.

He wasn’t upset about paying more.
He was upset that he didn’t see it coming.

This wasn’t about inventing a new fare system.
It was about reimagining one that already existed
— just without the confusion, the timing, and the shock.

Results

Launched in Bangalore for Auto

20%

Support tickets related to fare shock dropped significantly after launch

80%

Captains were more willing to accept rides in high-traffic zones

So, how did we do it?

Why does a necessary charge
feel unfair — just because it
came too late?

Defining the problem

For captains, the problem was simple

Longer rides in traffic meant more time, but not more earnings.
While other platforms already compensated captains for this, Rapido didn’t.

As a result, many captains avoided high-traffic areas altogether — especially during peak hours.

We knew that if we didn’t fix this, we’d eventually
- lower order acceptance
- longer wait times
- and a poor experience for everyone involved

For customers, the problem was different — but just as real

They weren’t frustrated by fare increases.
They were frustrated it was about paying without knowing why, or when it started adding up.

Every other platform made it part of the final bill.
We wanted to make it part of the journey.

5-6% of users face Time fare

≈ 200000 users daily

We had to answer two questions at once

Could we give captains what they deserved, without making riders feel blindsided?

And could we do it in a way that felt transparent, not transactional?

What we were betting on!

That customers would understand the change — if we told them clearly and early
That captains would accept more rides — if they knew they were fairly compensated
That we could reduce fare shock — simply by bringing users into the loop

Brainstorming

Everyone is involved from the

beginning

PMs, Developers, Product Designers were involved
to build the concept.
Building the concept together was key in uniting the
team: we had a common problem to solve.

Solution

It wasn’t about what we charged. It was about when — and how — we said it.

Simply being
Empathetic

1 Homescreen

2 Banner on FE screen indicating about heavy traffic*

3 Captain search screen

3 During ride when exceeding estimated ETA*

4 Payment details

An iterative process

It took us some time to land on

sometimes we were satisfied

with. Many user-tests and

rounds of feedback were necessary.

v3

v1

v2

Exploration

We’ve explored different narratives, component types and UI’s

Our approach to navigate

through deep water

Comic strip as a new visual style for bottomsheets & Goodwill narrative for copies

Visual approach

We chose comics over clean UI.
Everyone in India has grown up seeing comics — from Champak to Tintin.
They’re familiar, expressive, and make even serious topics feel light and clear.

Plan A: Comic Strip
Relatable, emotionally expressive
First-time in any ride-hailing app
️ Untested format — could’ve gone either way

Plan B: Minimal Illustration (Traffic Light)
✓ Visually calm and modern
❌ Too abstract, less context and emotional pull

Early user tests showed higher comprehension and more impressions with comics.
So we went bold — and it worked.

Copywriting approach

We chose empathy over instructions.
Instead of telling users what changed, we made them feel why it changed.

Plan A: Goodwill Approach (Empathy)
Built a connection with lines like:
“Heavy traffic means extra time on the road for your Captain.”
Framed the situation without blame
Made the fare feel fair

Plan B: Direct/Instructional
Clear and concise
❌ Felt cold, lacked emotional impact
❌ Didn’t reduce friction — only stated the fact


It led to less friction, fewer questions, and more user understanding.

A little extra helps us keep going!

Heavy traffic on the way

slows down the ride

Leading to Longer ride time

Comic style visual

Traffic delays are not ideal, so we make sure fare is minimal.

First 10 minutes free,

time fare starts only at 3:25 PM

If delayed further, it’s ₹0.5/min but it won’t exceed ₹20.

Empathetic Copy

Version 1 was launched

Continuous iterations are going on, ad we are soon launching it in all tier 1 cities for Auto & Cab both

How We Built Shared Auto for Everyday India

Shared mobility works offline. We just had to earn our spot in it. We focused less on onboarding, and more on belonging.

Results: 500+ rides everyday, 20% of first time shared auto users used Rapido once in 2 weeks

Platform: Android & IOS

Case study →