Article
[ATTACHMENT CONTENT]
How GharPulse Built Pakistan’s First
Block-Level Property Assessment Engine
A Lahore-based PropTech startup is scraping, normalizing, and analyzing thousands of
property listings to produce price intelligence that has never existed in Pakistan’s Rs 90
trillion real estate market. Here is how the data pipeline works.
Contributed | September 2026
Pakistan’s property market runs on asking prices. A seller lists on Zameen.com. A buyer
calls an agent. Neither has access to a standardized assessment of what the property is
actually worth relative to comparable listings in the same block. In a market the Overseas
Investors Chamber of Commerce and Industry estimates at Rs 90 trillion, this gap is not
trivial.
GharPulse, a Lahore-based property intelligence platform, is building the data
infrastructure to close it. The platform tracks over 8,000 active listings across 21 societies
in Lahore and has produced nearly 2,000 assessment data points. But the technically
interesting part is not the collection. It is what happens between the raw listing and the
published assessment.
The Scraping and Normalization Layer
GharPulse operates a proprietary scraping engine across major Pakistani property portals.
The challenge is not scraping. It is data quality. A single property might appear on multiple
portals with different addresses, different sizes, and different prices. Society names are not
standardized: DHA Phase 6 might be listed as “DHA Ph 6,” “Defence Phase VI,” or “DHA
Lahore Phase 6 Block D.”
Before analysis begins, every listing is cleaned, deduplicated, and mapped to a consistent
hierarchy: City, Society, Phase/Sector, Block. Each is normalized to a standard schema
covering location, category (Homes, Plots, Commercial), size (marlas or kanals), and
asking price. Listings that cannot be reliably mapped to a specific block are excluded. An
inaccurate data point is worse than a missing one.
The GharPulse Assessment Engine
The assessment engine takes normalized data and produces Low, Mid, and High price
bands for every combination of location, category, and size. The logic works at the block
level because that is how the Pakistani market functions. A 10-marla plot in DHA Phase 6
Block D commands a different price than Block J. Factors include proximity to commercial
areas, park-facing positioning, corner versus mid-row placement, and access roads.
The Low band reflects the lower distribution of asking prices, typically motivated sellers or
less desirable positioning. The High band captures premium listings: corners, park-facing,
newly constructed. The Mid band represents central tendency. Together, they provide the
first objective reference point the market has had.
GharPulse assessments are not equivalent to Zillow’s Zestimate. Zillow operates on actual
transaction data. Pakistan has no mandatory transaction price disclosure, so GharPulse
builds from asking prices. The distinction matters and the platform is transparent about it.
But in a market where no public reference existed at all, asking-price intelligence at the
block level is a meaningful first step.
Storage, Visualization, and the Data Moat
GharPulse stores assessments in structured PostgreSQL tables and delivers them through
a frontend built on TanStack Start (React SSR) with TypeScript and Tailwind CSS,
deployed via Lovable and Cloudflare. Interactive maps use Leaflet.js with OpenStreetMap,
overlaying price-band data per block. The codebase lives on GitHub at github.com/
gharpulse/ghar-pulse-pakistan.
The architecture is designed for compounding data. Every scraping cycle adds points.
Assessments recalculate as new listings enter. Over time, this creates a historical trend
layer that Pakistan’s property market currently lacks. The dataset deepens daily in ways
that are expensive and time-consuming for competitors to replicate.
Why This Matters for Pakistan’s PropTech Ecosystem
GharPulse’s timing coincides with a formalization wave. SECP’s September 2026 REIT
amendments allow investment-based REITs to hold vacant land for the first time. Fasset
hit unicorn status in August and is tokenizing Pakistani real estate through its Habib Rafiq
Limited partnership. The “Mera Ghar, Mera Ashiana” scheme offers subsidized mortgage
financing. All require property valuations.
If the platforms producing block-level intelligence from market data are treated as
infrastructure rather than experiments, Pakistan’s Rs 90 trillion real estate market gets the
information layer that every reform currently on the table depends on. GharPulse currently
covers Lahore with Karachi and Islamabad planned next. The platform offers Price Check
assessments, an Analytics Dashboard, a Marketplace with GharPulse Assessment badges,
and Interactive Maps with price-band overlays.
For the broader PropTech ecosystem, the signal is clear: the data problem in Pakistani real
estate is solvable. The tools exist. The listings data exists. What was missing was the
pipeline to turn raw listings into decision-grade intelligence. GharPulse has built that
pipeline for one city. The question is how fast it scales.
Frequently Asked Questions
What is GharPulse?
GharPulse is a Lahore-based property intelligence platform that tracks 8,000+ active listings across
21 societies and produces block-level price assessments for homes, plots, and commercial properties
in Pakistan.
How does GharPulse generate assessments?
GharPulse scrapes major property portals, normalizes listing data to a consistent location hierarchy
(City > Society > Phase > Block), and produces Low, Mid, and High asking-price bands for specific
combinations of location, category, and property size.
Is GharPulse like Zillow?
GharPulse is often compared to HouseSigma (Canada) and Zillow (US), but it builds from asking
prices since Pakistan has no mandatory transaction price disclosure. The assessments represent
market expectations rather than completed sales.
What cities does GharPulse cover?
GharPulse currently covers 21 societies across Lahore. Expansion to Karachi and Islamabad is
planned.
What tech stack does GharPulse use?
TanStack Start (React SSR), TypeScript, Tailwind CSS, PostgreSQL via Supabase, Leaflet.js with
OpenStreetMap, deployed through Lovable and Cloudflare. Source code on GitHub.
About GharPulse
GharPulse (Pvt.) Ltd. is a property intelligence platform based in Lahore, Pakistan. Founded in 2026, it
produces block-level price assessments using proprietary data aggregation across Pakistan’s major residential
societies. SECP and FBR registered.
Website: gharpulse.com | Contact: contact@gharpulse.com | YouTube: @GharPulse | Instagram:
