Research snapshot · August 14, 2026

Where does Rabbu get its data?

An evidence-based audit of Rabbu’s Airbnb analytics, Redfin property inventory, Zillow benchmarks, seller-contributed records, and Google integrations—separating what is proven from what is probable and what remains contractually unknown.

01 · Executive verdict

Airbnb is the primary STR source

Automated web collection is the strongest explanation, but private agreements cannot be ruled out.

01

High confidence

Airbnb → Rabbu analytics

Rabbu says it pulls market data directly from Airbnb, has gathered millions of US listings, and refreshes its calculations weekly.

Most likely mechanism: crawling/scraping, calendar snapshots, and proprietary transformation.

02

High confidence

Redfin → public sale inventory

Historical application code exposes Redfin property IDs, listing IDs, URLs, cached Redfin images, and an “Attribute synced from Redfin” label.

Exact mechanism remains unknown: scraper, approved feed, MLS route, or private agreement.

03

Declared source

Zillow ZHVI → home values

Rabbu’s 2026 market pages attribute the home-value input used in its ROI analysis to the Zillow Home Value Index.

This is a published third-party benchmark, separate from Redfin listing inventory.

04

User-contributed

Sellers and agents → exclusive inventory

Owners and agents submit off-market listings. Rabbu states that “Actual Financials” for active STR properties are seller-reported.

The public site does not document the verification standard, required evidence, or audit rate.

Critical distinction: “No public license found” does not mean “no license exists.” Source, collection mechanism, and legal permission are three separate claims.

1AirbnbSTR listings and calendars
2RedfinPublic for-sale inventory
3Seller / agentExclusive listings and financials
4Zillow ZHVIHome-value benchmarks
5Google + usersMaps, addresses, and telemetry

02 · Source-of-truth matrix

The data flows

Filter the matrix by provenance type. Confidence refers to sourcing, not contractual permission.

Data / product Upstream source Likely mechanism Public license Confidence
Airbnb listings and calendarsRates, availability, amenities, ratings, reviews, and images A Airbnb direct Automated collection, snapshots, and internal processing Not found High
ADR, occupancy, RevPANRevenue, seasonality, and projections R Rabbu analytics Algorithms over Airbnb comparable sets N/A for algorithm; input rights unknown High
Public properties for saleAttributes, images, source IDs, and listing URLs R Redfin Automated sync; exact path undisclosed Not found High
Exclusive / off-market listingsDescription, asking price, images, and contact details S Seller / agent Form, upload, and automatic Airbnb import Depends on uploader agreement High
“Actual Financials”Historical operating performance for a sale listing S Seller-reported Submission/upload; verification undocumented Not a dataset license Mixed
Home valuesBenchmark used in ROI Score Z Zillow ZHVI Zillow Research dataset Source attributed; exact commercial terms unpublished High
Addresses and mapsAutocomplete, geocoding, maps, and Places G Google Maps Platform Commercial Google APIs Use declared; private account terms unavailable High
User and usage dataAccounts, leads, forms, IP, device, location, and analytics R Rabbu first-party Voluntary submission and web telemetry Governed by Privacy Policy High

03 · Airbnb evidence trail

Why automated scraping is likely

The source is confirmed. The exact collector implementation is inferred from multiple independent signals.

  1. 01

    Direct first-party sourcing statement

    Rabbu says it pulls market data directly from Airbnb and updates it regularly.

    Rabbu guide ↗
  2. 02

    Millions of listings and weekly computation

    Rabbu says it gathered millions of US Airbnb listings and refreshes data and calculations weekly.

    Methodology ↗
  3. 03

    Archived forward-calendar observations

    The 2023 UI defined ADR and occupancy from posted rates and reserved nights over the next 30 days, and exposed comparable calendars.

    Archived app ↗
  4. 04

    Technical evidence in the application bundle

    The archived JavaScript includes Airbnb room links, listing IDs, comparable calendars, and JSON endpoints for market data.

    JS bundle ↗
  5. 05

    Manual process explicitly involves calendar scraping

    Rabbu describes manual underwriting as identifying comps and scraping their booking calendars, then positions its calculator as automating that analysis.

    Rabbu guide ↗
  6. 06

    No public Airbnb software-partner listing

    Rabbu did not appear in Airbnb’s public software-partner directory on the research date.

    Partner directory ↗

Inference boundary: the evidence does not reveal Rabbu’s collector code, proxy vendor, or any private agreement. A private Airbnb authorization or undisclosed intermediary remains possible.

Observed input

Rate + availability

Public listing attributes, forward prices, and calendar availability can be recorded in recurring snapshots.

Proprietary layer

Clean + seasonalize

Comparable selection, regional outlier removal, monthly aggregation, and proprietary scoring.

Displayed output

Estimate, not a ledger

Without host/PMS data, a blocked night can be mistaken for a guest reservation.

04 · Derived analytics

“Actual” performance is mostly modeled

Public calendar availability does not disclose whether the host blocked a night or a guest booked it.

Current methodology

Trailing-12-month market rollup

Current market pages say revenue is derived from trailing 12 months of active comparable Airbnb listings. Rabbu averages monthly RevPAN, removes regional outliers, and rolls the result into a market-level historical average.

Proprietary score

ROI Score weighting

Revenue-to-price ratio
40%
Occupancy stability
30%
Market growth trend
15%
Supply / demand balance
15%

Methodology caveat

Booked versus blocked

Unless Rabbu has host/PMS actuals or special Airbnb transaction access, market occupancy and revenue are calendar-and-price inferences—not accounting records of confirmed reservations.

Narrow exception

Seller-reported financials

Rabbu states that historical income for active STR sale listings is reported by the seller. It does not publish the documents required, the audit process, or the rate of independent verification.

About Rabbu ↗

05 · Property inventory

The public sale inventory came from Redfin

This source attribution is unusually direct. The legal and technical route is not.

A

Current Rabbu statement

Rabbu says the “public listing” tag can show every property listed on Redfin, paired with projected Airbnb financials.

Rabbu vs. Zillow ↗
B

Archived property payload

A historical Rabbu property page contains cached images hosted under ssl.cdn-redfin.com.

Property archive ↗
C

Archived application fields

The bundle contains redfin_property_id, redfin_listing_id, redfin_url, and the admin label “Attribute synced from Redfin.”

Marketplace bundle ↗

What this proves: an automated Redfin-to-Rabbu synchronization path existed. What it does not prove: whether the path used scraping, an approved feed, MLS rights, or a private agreement.

06 · Supporting sources

Seller data, Zillow, Google, and users

These inputs serve different product functions and should not be conflated.

Exclusive inventory

Owners and agents

Rabbu allows owners and agents to post a property even if it is not listed on Airbnb, Redfin, or Zillow. If an Airbnb listing exists, Rabbu says it can automatically import listing details and images.

That automatic import may itself be another web-collection flow; implementation and image permissions are not documented.

Sell My Airbnb ↗

Home-value benchmark

Zillow Home Value Index

Rabbu’s 2026 pages explicitly separate “Rabbu proprietary analytics” for STR performance from ZHVI for home values. Zillow publishes ZHVI at state, metro, city, county, ZIP, and neighborhood levels.

Maps and first-party telemetry

Google APIs and Rabbu users

Rabbu’s Privacy Policy declares Google Maps Platform APIs, including Maps and Places. It also describes voluntarily submitted personal data and automatically collected IP, device, browser, approximate location, referrer, and usage data, with possible sharing to business partners and subprocessors.

The archived frontend also loaded Google Maps JavaScript API with Places, supporting its role in address autocomplete and mapping.

Privacy Policy ↗

07 · Negative findings

What the research did not find

These are bounded negative findings, not proof that private arrangements do not exist.

Market positioning evidence

Rabbu presents itself as the free AirDNA alternative

Its comparison content describes AirDNA’s Airbnb and Vrbo metrics as AirDNA’s offering—not as Rabbu’s upstream source.

Live Airbnb data claim ↗

08 · Licensing risk

What must be confirmed contractually

Risk indicates the need for due diligence; it is not a legal conclusion.

High riskAirbnb

Automated collection is likely. Airbnb’s general terms prohibit bots, crawlers, and scrapers without authorization, while its API terms limit access to approved programs and scopes and prohibit use of undocumented Airbnb APIs.

Contract status: unknown
High riskRedfin

The synchronization path is proven. Redfin’s terms prohibit automated crawling and scraping unless prior express written permission is granted.

Redfin Terms of Use ↗ Contract status: unknown
Lower riskZillow ZHVI

Published research dataset with explicit Rabbu attribution. The exact commercial-use and attribution implementation should still be checked.

Provenance: confirmed
Contract-dependentSeller data

Image, financial-data, and republication rights depend on Rabbu’s agreement with the uploader and the uploader’s underlying rights.

Verification standard: unpublished
Likely standardGoogle APIs

Use is declared and resembles a normal commercial integration, but Rabbu’s specific account agreement is not public.

Integration: confirmed; private terms unavailable

Questions that close the due-diligence gap

  1. Does Rabbu collect Airbnb data itself, or purchase it from a vendor? If a vendor is used, who is it?
  2. Is there a written Airbnb API Program or other agreement covering market analytics, storage, derivation, and redistribution?
  3. How does Rabbu distinguish a host-blocked night from a guest reservation, and what percentage comes from host/PMS actuals?
  4. What are the true cadence, coverage, matching/deduplication process, and historical depth?
  5. Does Redfin inventory come from a scraper, Redfin/MLS feed, or another licensor? What image and republication rights apply?
  6. What evidence must a seller provide for “Actual Financials,” and who verifies it?
  7. What sources support STR rules, taxes, zoning, and property-level regulatory claims?
  8. Which ZHVI series is used, how are geographies mapped, and under what commercial terms?
  9. Do upstream contracts permit export, customer API access, AI training, or sublicensing?

A · Appendix — Gap review

Second-pass findings (August 14, 2026)

Independent verification pass using Parallel.ai (Search, Extract, Task, FindAll) and the Internet Archive. No core conclusion was overturned; the findings below refine, extend, or add to the report.

New context

Rabbu was a property manager (2016–2023)

Rabbu began as an STR property-management company in Charlotte, raised $1.5M in 2019, became one of the market's largest operators, and exited management in May 2023 — formally announcing its marketplace pivot in October 2025. It therefore held first-party PMS/booking actuals for its own portfolio, but only through 2023 and centered on Charlotte. Post-2023, "actual booking performance" cannot rest on its own operations.

Stronger evidence

Redfin sync still live in July 2025

A market page archived July 11, 2025 hydrates source_properties_from_redfin: true, serves cached photos from ssl.cdn-redfin.com, and separates synced inventory via is_seller_listing: false. The report's 2022–2023 archive evidence extends to mid-2025; the August 2026 state is unverifiable (Cloudflare).

Archived page ↗

Vendor viability

~4 employees, seed-stage

Tracxn shows 4 employees (Dec 2023 and Dec 2024) and roughly $1.9M total funding. A team this small operating "1.1M+ active listings" implies either very lean infrastructure or outsourced collection — and real counterparty risk in any enterprise negotiation.

Tracxn ↗

New surface

Chrome extension covers Compass too

The official "Rabbu Airbnb Calculator" extension (~1,000 users) injects a "View on Rabbu" button on Zillow, Redfin, and Compass listing pages — Compass is an upstream surface absent from the report. It ties third-party listing inventory directly into Rabbu's estimate funnel.

Chrome Web Store ↗

Coverage claim

Datarade: 1-year history only

An unclaimed Datarade profile lists 2.14M total listings, 342M nights tracked, weekly updates, and just 1 year of historical data — the most concrete coverage claim found, and a major limitation versus AirDNA for multi-year seasonality. Not evidence of an active data-licensing business.

Datarade profile ↗

Monetization confirmed

Lead-gen with public pricing

Agents pay $150 per exclusive lead with a 20-lead minimum; lenders compensate Rabbu for qualified referrals (Total Quality Lending, theLender, Host Financial, The One Brokerage, plus the April 2025 Beeline partnership). Hydration data exposes the machinery: lead_types, lender_partners, CallRail tracking numbers, and a "quizard" qualification quiz. Free data is the bait; leads are the business.

Measured freshness (archived page, July 11, 2025)

ComponentTimestamp in pageAge at snapshot
Property cacheper-listing cache_updated_at"about 6 hours ago" / "1 day ago"Hours to 1 day
Airbnb market dataairbnb_last_updated_at2025-07-08~3 days (consistent with weekly refresh)
Zillow ZHVIzhvi_updated_at2025-06-13~4 weeks (consistent with monthly release)

One page, three ages: the same market page mixes hours-old inventory cache, days-old STR metrics, and month-old ZHVI. "Real-time data updated weekly" is accurate only for the STR component.

FindAll vendor landscape (18 matched)

A whole market sells scraped Airbnb data

A Parallel.ai FindAll run confirmed a mature US STR-data market: licensable analytics providers (AirDNA, Key Data, Airbtics, AirROI, Mashvisor, AllTheRooms, Transparent/Lighthouse, Beyond, Rentals United, Deckard) and turnkey scraping vendors with dedicated Airbnb products (Bright Data, Oxylabs, Actowiz, TravelScrape). Two implications: a 4-person Rabbu could plausibly buy scraped feeds rather than run its own collector — shifting due-diligence question #1 to "who executes collection and who carries the ToS risk?" — and genuine licensable alternatives exist for any product that needs STR data under contract.

Updated verdict for our use case: nothing found improves Rabbu's fit as a data supplier for a public-facing product — no public licensing framework, ~1 year of history, tiny team, and unchanged upstream risk. Realistic role: product benchmark and inspiration, not a data vendor. Full details in the research notes appendix.

09 · Full reference set

Primary evidence and relevant links

All references from the underlying research are included below. Access checked August 14, 2026.

Rabbu: product, methodology, and disclosures

Rabbu: current market methodology examples

Archived technical evidence

Upstream platforms and licensing context

Research method

The investigation covered Rabbu’s website, blog, legal pages, market pages, robots/sitemap behavior, historical data.rabbu.com and buy.rabbu.com subdomains, archived pages and JavaScript bundles, public GitHub results, press, and community discussions. No public GitHub repository attributable to Rabbu exposed the collector. Reddit and third-party directories were used only as leads, not as the basis for core conclusions.

This is a provenance and risk assessment, not legal advice.