Home > > > MagicBricks vs 99acres vs Housing: Listing Data Study 2026

MagicBricks vs 99acres vs Housing: Listing Data Study 2026

Author : Actowiz Solution | Published On : 04 Aug 2026

Introduction

Actowiz analyzed 200,000+ listings across MagicBricks, 99acres, and Housing.com in 8 Indian cities. Findings: the same property appeared on multiple portals with asking-price differences in 32% of matched cases (median gap ₹X lakh / Y%); duplicate and stale listings inflated raw inventory counts by 32%; and portal coverage skews sharply by city — no single portal gives a complete market picture anywhere.

The Problem With Reading One Portal

India's property portals are marketing channels, not registries. Brokers cross-post with different prices, listings outlive actual availability, and each portal's broker network skews coverage by city and segment. Anyone using portal data — buyers, proptechs, lenders, researchers — needs to know how big these distortions are. We measured them.

Methodology

ParameterCoveragePortalsMagicBricks, 99acres, Housing.comCitiesMumbai, Delhi NCR, Bengaluru, Pune, Hyderabad, Chennai, Ahmedabad, KolkataListings analyzed200,000+ (resale + rental)MatchingProject + tower/locality + configuration + area-band entity resolutionWindow60 days, weekly snapshotsFieldsAsking price, area, configuration, project, lister type, posting/refresh dates, photo count, RERA tag

  • Property Portals Covered: MagicBricks, 99acres, and Housing.com.

  • Cities Covered: Mumbai, Delhi NCR, Bengaluru, Pune, Hyderabad, Chennai, Ahmedabad, and Kolkata.

  • Listings Analyzed: 200,000+ property listings across resale and rental markets.

  • Matching Methodology: Entity resolution based on project, tower/locality, property configuration, and area band to identify comparable listings.

  • Monitoring Window: 60 days with weekly snapshots.

  • Data Fields Captured: Asking price, property area, configuration, project name, lister type, posting and refresh dates, photo count, and RERA tag.

Finding 1: Same Property, Different Price

Among cross-portal matched listings, 48% showed asking-price differences — typically the same broker anchoring differently per portal audience. Gaps ran widest in Mumbai resale (median 22%). For valuation models, this is noise that must be reconciled; for buyers' products, it's a feature worth surfacing.

Finding 2: Inventory Inflation — Duplicates & Stale Listings

  • Within-portal duplicates (same unit, multiple brokers): 42% of raw listings.

  • Stale listings (no refresh in 45+ days): 24%, concentrated in rental segments.

  • Net effect: raw counts overstate true available inventory by ~X% — any supply analysis on unclean portal data inherits this inflation.

Finding 3: Coverage Is City-Skewed

 

  • Mumbai: Strongest Portal (Listings Share): Portal, X%.

  • Bengaluru: Strongest Portal (Listings Share): Portal, X%.

  • Delhi NCR: Strongest Portal (Listings Share): Portal, X%.

Broker-network strength drives this — meaning multi-portal aggregation isn't optional for full-market coverage, it's the method.

The RERA Cross-Check

Joining listings to our state RERA datasets exposes the gap between advertised and registered reality: 38% of new-project listings carried RERA tags; 22% of those tags failed validation against the registry. For lenders and compliance teams, that join is the diligence product.

What Teams Do With This Data

  • Proptech & brokerages: deduplicated, multi-portal listing feeds with price-reconciliation fields.

  • Lenders & valuers: asking-price distributions by micro-market, cleaned of duplicates and stale stock.

  • Investors & researchers: supply and pricing trend series with documented cleaning methodology.

FAQs

How do you match the same property across portals?

Entity resolution on project + tower/locality + configuration + area band + lister fingerprints. Conservative thresholds mean reported cross-portal gaps come from high-confidence matches only.

How do you detect stale listings?

Refresh-date tracking plus content-change detection across weekly snapshots: listings with no movement past a threshold are flagged stale, with the threshold configurable per use case.

Can I get deduplicated inventory counts by locality?

Yes — cleaned supply series by micro-market are a standard deliverable, alongside the raw data and the dedup mapping for auditability.

Does this integrate with RERA data?

Yes — listings join our state-wise RERA datasets for registration validation, promoter history, and advertised-vs-registered analysis.