# UK Estate Agency Meta Advertising: Methodology and evidence register (2026)

**Publisher:** RealAdBook · Author: Oleh Odukalets · Release v1.1, 31 October 2026 (v1.0 published the same day; see Change log). **Fieldwork:** 30 October 2026. **Study type:** a purposively selected cross-sectional, one-day descriptive audit. **Status:** published evidence release; limitations and the incomplete review of two advertisers disclosed below.

## Question and scope
Among a defined set of UK estate agency brands appearing in the Best Estate Agent Guide 2026, which could be matched to active UK Meta advertising on 30 October 2026, and what purposes could be identified in the reviewed advertisements? This is **not** a survey of all UK estate agencies, a market share estimate, or a study of advertising effectiveness.

## Sampling and deduplication
The sampling frame was assembled on 30 October 2026 from Best Estate Agent Guide 2026 listings within a three-mile radius of the city centres of **London, Manchester, Birmingham, Bristol, Edinburgh and Glasgow**. This selection yielded **176 branch/listing appearances**, deduplicated to **108 estate agency brands**. Each brand was coded once even if it appeared in more than one sampled location, and the field `markets` lists all cities where that brand was found. The `market` field in the Ad Library file is a single nominal home/assignment label, not a local ad audience or the only city in which an agency operates.

Best Estate Agent Guide is a *filtered, performance-selected list*, compiled using evidence from TwentyEA and other signals. It does not produce a random sample of the entire UK market. Original publisher: https://bestestateagentguide.co.uk/about/research-methodology/ and https://bestestateagentguide.co.uk/about/rating-system/.

## Meta Ad Library search
Meta Ad Library was searched with **United Kingdom** country, **active** status and **all ads** media/ad-type settings on 30 October 2026. Businesses were searched by advertiser name, and matching Facebook pages were checked against the brand name and, where possible, the advertised destination website. For chains and franchises, matching branch or sub-brand pages were also considered. **The Meta search was UK-wide**: an agency's inclusion in a city's three-mile BEAG search does not establish that its adverts targeted that city. No impressions, ordinary commercial advertising expenditure, conversion metrics, lead costs or verified financial return are present.

### Advertiser status
- **A**: active ad cards found on an identified matching brand page.
- **AB**: active ad cards found on identified matching branch, franchise or sub-brand page(s), not the main page.
- **Z**: a matching Facebook page was identified but did not display active GB ads at the snapshot.
- **W**: only wrong/ambiguous same-name candidates, not reliably identifiable as the business.
- **N**: no matching Facebook page established in the review. **Neither N nor W means that the business has no Facebook page or never advertises**.

## Counting and incompleteness
**30/108 brands** were coded A or AB; **46/108** coded Z; **29/108** N and **3/108** W. In the 30 active-brand records there were **275 active ads recorded** (the Ad Library results count). The library shows ads that share the same creative as one card, so card and ad counts differ: **201 cards were read, covering 240 of the 275 ads**. Two active brands were partly reviewed: Your Move (29 cards covering 30 of 64 ads) and Red Loft (12 of 13 ads). Their 35 unreviewed ads may contain categories that the review did not record. **The remaining 28 active brands have all 198 recorded ads reviewed**; for example, Winkworth's 11 cards cover all 37 of its ads, Yopa's 3 cover 12 and Coulters' 9 cover 12.

Fields: `active_ads` is the number of active ads the library reported for the brand's matched page(s); `ad_cards_read` is the number of cards read; `ads_reviewed` is the number of ads those cards cover. A card is not necessarily a unique creative concept, and no unique-creative count is available. Counts by advertising purpose are of *brands in which the code was found*, not counts of campaigns or enquiries. Codes are multiple-response, so values do not add to 100%. Full review status is available per brand in the CSV, permitting analysis of the 28 completely reviewed advertisers separately.

### Advertising purpose codes
- **S**: seller or valuation acquisition.
- **I**: instant online valuation specifically promoted in the ad.
- **P**: numeric or ranked evidence used in a seller acquisition ad (e.g. claimed % of asking price).
- **B**: for-sale listings.
- **N**: new homes or developments.
- **R**: to-let listings.
- **L**: landlord acquisition.
- **C**: market guide, report or resource download.
- **F**: brand, page-follow or community promotion.
- **J**: recruitment.

Codes represent a positive finding from the reviewed material. No code does not establish that a business never uses the approach, particularly when `ads_reviewed < active_ads`. The CSV notes are fieldwork annotations, not independently proven advertising outcomes. Original `library_url` values can be followed to revisit identified pages. The library is live and historical display conditions can change after the snapshot.

## Parallel website audit (not part of the current quantitative findings)
A website coding protocol separately considered buyer and seller journeys, valuation, landlord services, trust evidence, fees, and positioning for the same 108 brands. At the fieldwork cutoff 54 sites were marked read, eight read but identity unverified, 15 were blocked to the automated collector, 20 not found, eight returned wrong-site candidates, two did not render, and one was found but not read. The 62 provisional coded rows were generated in an automatic first pass and **have not passed manual identity/variable verification**. They are not used for published website statistics. The UK region also spans distinct legal systems; no tenancy/fee compliance finding is made in this release.

## Reproduction and corrections
The public downloads below preserve both the 108-brand eligibility frame and the 108-brand Meta coding notes, with a common `id` (UK001–UK108) for joining. The evidence files contain page IDs / Ad Library URLs where available. Limitations: no complete per-card ID/screenshot archive, no independently validated second coder, and no validation of ads that Meta no longer displayed in its live interface. A subsequent revision should complete the two partial reviews, verify N/W identities and publish a full ad-level archive and website recoding. Corrected findings must use a new revision date and a change log. Contact: https://realadbook.com/contacts.

## Data files
- [Meta brand coding, 108 records (CSV)](/research-data/uk-estate-agency-meta-ad-library-2026.csv)
- [Panel selection and website collection statuses (CSV)](/research-data/uk-estate-agency-panel-2026.csv)

**Reuse:** The original RealAdBook coding/data compilation is supplied for reproducibility under CC BY 4.0 with attribution and canonical report URL. The underlying brands, logos, advertisements and proprietary third-party materials remain the property of their respective owners. This is an independent study; it is not endorsed by Meta, the Best Estate Agent Guide, Property Academy or TwentyEA.

**Commercial-interest disclosure:** RealAdBook also sells real estate marketing educational and Market Intelligence products. This study uses publicly visible information and does not use client-paid campaign data.

## Change log
- **v1.1 (31 October 2026):** v1.0 treated each card that grouped several ads as one read ad and reported 201 of 275 ads reviewed, with five advertisers partly unread. Counting the ads each card covers gives 240 of 275 reviewed, with two advertisers partly unread (Your Move, Red Loft). The CSV column `ads_read` is renamed `ad_cards_read` and a new `ads_reviewed` column is added. Brand-level purpose counts are unchanged.
- **v1.0 (31 October 2026):** first release.
