Building a Comparison Platform for India

"The Market Is Too Crowded." They Were Right — and Wrong.

Every investor meeting in 2021 started the same way: "BankBazaar is already doing this. PolicyBazaar is already doing this. What's your differentiation?" The question was fair. The implied advice — don't build it — was wrong.

Crowded markets aren't uniformly served markets. BankBazaar was dominant but desktop-first, and genuinely terrible on mobile. PolicyBazaar owned insurance but had minimal presence in the lending comparison space. The "crowded" market had significant whitespace — it just wasn't visible from the incumbent's perspective.

The strategic insight wasn't "this market is wide open." It was: "this market has been served by incumbents optimizing for their original acquisition patterns — desktop, English-speaking, financially literate. A huge segment of the market is underserved by those patterns."

The Market Reality

What the Incumbents Were Missing

We ran a 3-month competitive analysis — not just product features but user experience, language support, mobile behavior, and the segments each platform was built for.

Platform Gap	User Segment Affected	Our Opportunity
Desktop-first UX on mobile	65% of Indian internet users (mobile-only)	Mobile-native design as baseline
English-only interface	Hindi and regional language speakers	Hindi + 4 regional languages from launch
Financial jargon throughout	First-time borrowers / insurance buyers	Plain language, in-context explanations
No EMI calculator in comparison	EMI-sensitive borrowers	EMI-first display, not APR-first
No offline-adjacent features	Users with variable connectivity	Comparison caching for offline review

The gaps weren't random. They reflected the incumbents' founding context — built in 2008–2013, when India's internet user base was heavily urban, English-speaking, and desktop-centric. The market had moved; the products hadn't fully moved with it.

Our User

Building for the User the Incumbents Weren't Building For

We did 40 user interviews in Tier 1 and Tier 2 cities. The pattern that emerged defined our entire product approach.

User Profile	Current Behavior	Pain Point	What They Needed
Tier-2 city first-time borrower	Walks into bank branch	No online option that speaks his language	Hindi UX + EMI-first comparison
Mid-20s mobile-only user	Googles, opens 4 apps	No unified comparison; too much jargon	One-screen comparison, plain language
Self-employed professional	Compares manually, asks CA	Products assume salaried income	NACH/alternative income acknowledgment
First insurance buyer	Relies on agent recommendation	Doesn't understand policy terms	Term-by-term plain language breakdown

Every profile we interviewed had some version of the same complaint: "I don't understand what they're asking me." Not confusion about which product to choose — confusion about what the words meant. Clarity was the product, not just a feature.

Strategy

The Product Strategy That Unlocked the Market

The strategic frame wasn't "build a better BankBazaar." It was "build the first comparison platform that actually works for someone who's never done this before, on a phone, possibly in Hindi."

Incumbents optimize for the user who already understands the category. The bigger opportunity is always the user who doesn't understand it yet — because that's where the market is actually growing.

Three non-negotiable design principles: (1) EMI before APR — show the number users think about, not the number that's technically more accurate; (2) Every term must be explainable in one sentence of plain language; (3) Mobile-first meant the desktop version was built second, not adapted from desktop.

Product Architecture

What Differentiated the Platform

EMI-First Comparison Architecture

Interest rate displayed second; monthly EMI displayed first and prominently. Comparison sorted by EMI by default, not APR. Users make decisions in EMI terms — the interface respects that reality rather than correcting it.

Multi-Language Interface

Hindi at launch; Tamil, Telugu, Marathi, Kannada by Month 4. Language detection from device locale; manual toggle available. Not translation — native copy written in each language, not translated from English.

Plain-Language Term Engine

Every financial term in the platform has a one-sentence plain-language equivalent stored. Tooltip on tap (mobile) or hover (desktop) shows the explanation. Built as a content system — 400+ term definitions, maintained separately from product code.

Eligibility Pre-Check

Before showing full comparison, 4-question eligibility pre-check that surfaces only products the user is likely to qualify for. Reduces comparison overwhelm; increases application completion rate by 3.8×.

Technical Build

What Was Hard to Build

Multi-Language Content System

→Content management system built to handle 5 languages simultaneously
→All UI strings externalized — no hardcoded English anywhere in the codebase
→Translator partnership: native speakers in each language, not translation APIs
→Launch QA in each language required native-speaker review, not just functional testing

Eligibility Engine

→4-question pre-check maps to 40+ lender eligibility criteria under the hood
→Rules engine maintained by credit team, not engineering — configurable without code deploy
→Output: 'likely eligible,' 'possibly eligible,' 'likely not eligible' per product (never hard no)
→Lender criteria updated monthly; rules versioned and auditable

Mobile Performance

→Target: LCP < 2.5 seconds on 4G; <4 seconds on 3G (prevalent in Tier 2)
→Comparison data cached after first load; offline browsing of cached comparison supported
→Images: WebP with AVIF fallback; all optimized to sub-30KB per product card
→Critical path: full comparison rendered in <1.2 seconds from tap to data visible

Traction

18 Months of Building and Learning

Month 1–4 · Foundation

Core Product + Hindi Launch

→ First 12,000 users. Hindi-speaking users: 44% of signups (vs 15% industry benchmark)

✓Core loan comparison: personal loan, home loan, credit card
✓Hindi language version at parity with English
✓EMI-first comparison architecture live

→ First 12,000 users. Hindi-speaking users: 44% of signups (vs 15% industry benchmark)

Month 5–9 · Regional Languages

4 More Languages + Insurance

→ 60,000 MAU. Regional language users: 38% of total.

✓Tamil, Telugu, Marathi, Kannada added
✓Insurance comparison launched (term life, health, motor)
✓Eligibility pre-check live for loan products

→ 60,000 MAU. Regional language users: 38% of total.

Month 10–18 · Scale

Tier 2 + Product Depth

→ 180K MAU. Application completion rate: 3.8× incumbents. CAC: 40% lower via SEO.

✓SEO content in 5 languages — drove 48% of acquisition
✓Offline comparison caching launched (key for Tier 2 connectivity)
✓180,000 MAU; application completion 3.8× industry benchmark

→ 180K MAU. Application completion rate: 3.8× incumbents. CAC: 40% lower via SEO.

Results

At Month 18

180K

Monthly Active Users

18 months from zero

3.8×

Application Completion

vs incumbent average

Languages

from day one for Hindi

−40%

CAC

vs incumbent benchmark (SEO-led)

Insights

What Building in a "Crowded" Market Taught Me

Crowded markets have uncrowded segments

BankBazaar and PolicyBazaar are dominant in the English-first, financially-literate, urban segment. They had minimal presence in Hindi-speaking Tier 2 cities. The market looked crowded from the outside; from the inside, the specific segment we targeted was wide open.

Language is a product decision, not a localization decision

Adding Hindi wasn't translating the English product. It was rebuilding the product from first principles for a user whose context, vocabulary, and decision process were different. Translated products feel translated. Native products feel native.

Application completion rate is the real platform conversion metric

Generating comparisons is table stakes. What actually matters is whether users complete the application on the lender's platform. Our 3.8× industry-benchmark completion rate came from eligibility pre-screening — users who applied were users who'd already pre-qualified.

Failures

What We Got Wrong Early

✕ First Hindi version was a translation, not a localization

Why it broke

We translated the English UI copy into Hindi using a freelance translator. The result was grammatically correct but felt unnatural — financial Hindi as written, not spoken. User testing feedback: 'It sounds like a bank circular.'

What fixed it

Hired native Hindi speakers to rewrite all copy from scratch, using conversational Hindi. The rewrite took 6 weeks but produced copy that felt like a trusted friend explaining finance — which was the goal.

✕ EMI calculator was accurate but confusing

Why it broke

First version showed annual EMI, monthly EMI, total interest payable — all on one screen. Too much information. Users with lower financial literacy were more confused after seeing the calculator than before.

What fixed it

Default: only monthly EMI and total repayment shown. 'Show more' expands to full breakdown for users who want it. Simplicity for the majority; depth for the minority.

Future

Where This Goes

Personalized Comparison Ranking

Rank comparison results by individual user's predicted approval probability, not just rate. A lender who rarely approves users with the applicant's profile moves down the list — even if their rate is better.

Credit Score Advisory

For users who don't qualify for preferred products, a 90-day credit improvement plan — what to do to reach the next tier of products. Retention mechanism and trust builder simultaneously.

Embedded Finance

Move from comparison to facilitation — completing the application in-platform, not redirecting to lender. Remove the last point of drop-off.

"The market is too crowded" is rarely the full story. It's too crowded for the obvious approach. The real question is: for which specific user, in which specific language, with which specific context, does the right product not yet exist?

Building a Real-Time Cab Fare Comparison Engine (Uber vs Ola vs InDrive)

Product

Creating a Scalable Notification System for 1M+ Users (Trigger-Based)

Product

Building a CRM That Handles 180K+ Monthly Transactions at 99.6% Uptime

12 min read

Product

How We Built a Multi-Calculator Finance App That Drives Daily Retention

8 min read

Product

Building a Real-Time Cab Fare Comparison Engine (Uber vs Ola vs InDrive)

11 min read

All Case Studies

Browse all 30 breakdowns

Work With Me

Let's build something together

Connect on LinkedIn

Follow for more insights

Where to go next

Fintech Product Consultant
The consulting work this sits inside, with the rest of the evidence behind it.
Day 30 retention benchmark, fintech & banking
Published figures, traced to their sources — and where the sources disagree.
Day 1 retention benchmark, fintech & banking
Published figures, traced to their sources — and where the sources disagree.
Product Growth Score
Thirteen questions that locate which part of the funnel is actually constraining growth.

Designing a Comparison Platform Like Monito for the Indian Market

Results at a glance

"The Market Is Too Crowded." They Were Right — and Wrong.

What the Incumbents Were Missing

Building for the User the Incumbents Weren't Building For

The Product Strategy That Unlocked the Market

What Differentiated the Platform

What Was Hard to Build

18 Months of Building and Learning

At Month 18

What Building in a "Crowded" Market Taught Me

Crowded markets have uncrowded segments

Language is a product decision, not a localization decision

Application completion rate is the real platform conversion metric

What We Got Wrong Early

Where This Goes

You might also like

Where to go next

"The Market Is Too Crowded." They Were Right — and Wrong.

What the Incumbents Were Missing

Building for the User the Incumbents Weren't Building For

The Product Strategy That Unlocked the Market

What Differentiated the Platform

What Was Hard to Build

18 Months of Building and Learning

At Month 18

What Building in a "Crowded" Market Taught Me

Crowded markets have uncrowded segments

Language is a product decision, not a localization decision

Application completion rate is the real platform conversion metric

What We Got Wrong Early

Where This Goes