04 Product Analysis · Funnel Analytics

NoBroker

The paywall is where trust breaks down — or is it just where it becomes visible?

Out of every 100 users who entered NoBroker's rental funnel, only 2 converted. Using SQL, Python, and firsthand product teardown, I analyzed 50,000 users across 468,000+ events to identify where trust breaks, why users drop off, and which single change could unlock the highest revenue impact.

Role
Product Analyst (Case Study)
Dataset
50,000 users · 468,959 events · Jan–Mar 2026
Methods
Product Teardown · Funnel Analysis · SQL · Python · Experiment Design
Tools
SQL Python Power BI Jupyter Figma
The Question

Why do users abandon NoBroker at the exact moment they seem most interested?

First glance
The obvious answer was wrong.

At first glance, the answer looked obvious. The paywall. 85% of users who reached it never paid.

But funnels rarely break where they appear to. The goal of this analysis was to understand where trust actually breaks — and whether the paywall was the cause or merely where the damage became visible.

Product Teardown
Before touching any data, I used NoBroker as a real renter.

I wanted to understand what users actually see, what NoBroker asks for, where friction appears, and how the platform monetizes. During this walkthrough, I discovered something unexpected.

The signup wall appeared at two completely different moments — sometimes while browsing, sometimes only after clicking a listing. This single observation changed the direction of the entire analysis.

Path A — Discovery Interrupted
Mid-Browse Wall
Wall interrupts before intent forms
Browse listings
Signup wall appears
Interrupted. Cold.
Path B — Intent Interrupted
Post-Intent Wall
Wall triggers after a desired listing found
Browse listings
Find a desired listing
Signup wall appears
Motivated. Context intact.
Key Observation
Same wall. Different user mindset. That difference became critical.
Funnel Analysis
Only 2% of users became paying customers.

Three cliffs emerged. The question wasn't where users left — it was why.

NoBroker Rental Funnel — Bangalore, Jan–Mar 2026
50,000 users → 1,033 paid  |  2.07% end-to-end conversion
01
Visitors leaving immediately
47.8% of users who landed never searched. They hit the page and left without engaging further.
02
Users abandoning the signup wall
42.1% dropped at signup completion — the largest single fixable drop in the entire funnel.
03
Users refusing to pay
85% of users who saw the paywall never converted — even after unlocking contact details.
Path Analysis
This is where the case becomes interesting.

Although both groups saw the same signup wall, their behavior was dramatically different. Users who encountered the wall after finding a desired listing converted significantly better than users interrupted while still browsing.

Path A vs Path B — Funnel Comparison
Path A: wall interrupts browse  |  Path B: wall triggered by contact intent
The shift
The wall itself wasn't the problem. The timing was. That shifted the investigation from "how do we improve the signup form?" to "should this wall exist here at all?"
Paywall Investigation
At first the paywall looked guilty. Then it didn't.

Most users who saw it left. But digging deeper revealed something surprising: nearly one-third of users returned multiple times before deciding.

They weren't reacting impulsively. They were evaluating. And still choosing not to pay. That suggested the paywall wasn't creating distrust — it was revealing distrust that already existed.

Paywall Hesitation — Users Who Returned But Didn't Convert
27 min avg difference between converters and non-converters
The Insight

The paywall is where users express broken trust. Not where trust breaks.

Every step before payment weakened confidence. The paywall simply became the first moment users could say no.

Root Cause
Not one broken screen. Five compounding problems.

Rather than a single friction point, the problem was cumulative. Each stage compounded the previous one. The final drop was simply where accumulated frustration surfaced.

1
Stage 01
Expectation Gap
Users arrive expecting a straightforward rental search. The product's monetization model isn't surfaced upfront, creating a dissonance between expectation and reality.
2
Stage 02
Poor Listing Quality
Questionable or incomplete listings early in the browse experience reduce confidence that the platform has genuine, high-quality inventory worth pursuing.
3
Stage 03
Premature Signup Wall
Appearing before users have found anything worth pursuing, the wall signals that the product doesn't trust the user yet — before the user has had reason to trust the product.
4
Stage 04
Undisclosed Phone Exposure
Contact unlock reveals the user's phone number to an owner without explicit warning. Users learn about this exposure only after they've committed, triggering immediate regret.
5
Stage 05
Confusing Paywall
By the time users reach payment, trust is already damaged across four prior touchpoints. The paywall becomes the first moment they can vote with their feet — and most do.
Prioritization
Five interventions. One highest-leverage move.

Instead of prioritizing by effort alone, I scored each intervention against impact on trust, revenue potential, engineering complexity, and speed to validation.

Intervention Trust Impact Revenue Potential Complexity Speed to Validate
Remove mid-browse signup wall ★
Disclose phone exposure before unlock
Improve listing quality signals
Reframe paywall value proposition
Reduce spam from unlocked contacts
★ Recommended highest-leverage intervention  ·  Red dots = severity  ·  Green dots = low complexity / fast
Highest-Leverage Recommendation

Remove the mid-browse signup wall.

Why this
Attacks the largest irreversible drop in the funnel. Users who leave at this stage never return.
Unlike other fixes
Listing quality and pricing take months. This can be validated through a simple A/B experiment in days.
The mechanism
Delay signup to the moment of intent — when a user clicks to reveal owner details. Higher motivation means higher completion.
Experiment Design
The signup wall test.

A single well-designed experiment can validate the highest-leverage recommendation without committing to a full rollout.

Control
Current experience
Signup wall appears mid-browse, before any listing-level intent is established. User must sign up to continue browsing.
Variant
Intent-triggered signup
Signup wall delayed until the user clicks "Get Owner Details" on a specific listing. Browsing remains open. Wall only appears when contact intent is demonstrated.
Primary Metrics
Signup completion rate
Contact unlock rate
Paywall conversion rate
Guardrails
Total registered users ≥ control
Session length not reduced
No increase in spam reports
Projected Impact
Users recovered. Then revenue.

Based on modeled recovery scenarios from the 42.1% signup cliff. Even conservative recovery unlocks meaningful revenue impact at NoBroker's scale.

Conservative
+10% recovery
Expected
+30% recovery
Optimistic
+50% recovery
Recovery of users lost at the signup cliff (15,000+ users in dataset period)
Modeled Revenue Impact (Expected Scenario)
Users recovered (monthly)
~4,500
Est. additional conversions at 2% rate
~90
At avg. plan value
Meaningful revenue uplift per month without product changes — purely from funnel timing adjustment.
By the numbers
What the data revealed.
End-to-End Conversion
2%
50,000 users entered the funnel. Only 1,033 paid.
Largest Drop — Signup
42%
Of users who hit the signup wall abandoned at completion. The largest product-fixable cliff.
Path B Signup Advantage
+19pp
Intent-triggered wall (69.8%) vs mid-browse wall (50.0%) signup completion.
Users who returned to the paywall multiple times converted at nearly identical timings to those who didn't — meaning time wasn't the barrier. Trust was. No amount of waiting was going to fix a confidence problem baked into earlier touchpoints.
What I Learned
Funnels show where users leave. They don't explain why.
The step before the drop matters more than the drop
The paywall initially looked like the problem because that's where the largest visible number appeared. But the real breakage happened at four earlier touchpoints. The most valuable investigation was always one step upstream.
Same friction, different context, different outcome
Path A and Path B users saw an identical signup wall. The difference in behavior was entirely explained by what happened before. This confirmed that product decisions can't be evaluated in isolation — context determines meaning.
Users rarely abandon for one reason
Trust eroded across five stages before the paywall. No single intervention would have solved it. The highest-leverage move was the one that addressed the largest irreversible drop — not the most obvious symptom at the end.