An audit of 430 active Facebook and Instagram ads from 54 advertisers, collected on 30 September 2026
RealAdBook collected every active Meta ad we could find for property in Marbella and coded what each one tells a buyer before the click. Most of these ads sell the feeling of living on the coast very well, while far fewer tell the buyer what the property costs, who it is meant for or why it is worth a call.
Between the sea view and the contact form, a Meta ad has a few seconds to tell a buyer three things, which are what the property costs, whether it is meant for someone like them and why it deserves a conversation. We wanted to know how often active Marbella ads actually do this, so we collected 430 of them from the Meta Ad Library on a single day and coded the 255 unique creatives behind them against a published codebook.
Marbella is one of the most international housing markets in Spain. Notary records show 4,379 home sales in the municipality in 2025, about 8% fewer than in 2024, with 62% of the purchases made by foreign buyers and the largest groups coming from the United Kingdom, Sweden, the Netherlands, Poland and Germany1. The same data puts the average sale at €736,866 and €4,424 per square metre, while idealista recorded a record asking price of €5,956 per square metre in August 20261,2. Across Málaga province the foreign share of purchases reached 37% in the second quarter of 2026 according to the Colegio de Registradores3.
A market where most buyers live abroad and the average sale is worth more than €700,000 is exactly where Meta advertising becomes attractive, because an agency in Marbella can reach a family in Stockholm or an investor in Amsterdam in their own language for a fraction of what a portal listing costs. It is also where the cost of a poorly matched enquiry is high, because every conversation with the wrong buyer takes time that a broker could have spent with the right one.
A lot is written about how to run property ads on Meta, but very little is measured about what the market actually says in them. Agencies usually see their own ads and a few competitors, and journalists usually see the finished campaigns that someone chose to show them. The Meta Ad Library makes every active ad public, which makes it possible to read the whole visible market on one day and count what it tells a buyer.
The aim of this study is to describe, with counts that anyone can check, how property in Marbella is presented to buyers in active Meta advertising, and to show which pieces of information a buyer receives before deciding whether to leave their contact details. The study answers five questions.
| Element | Definition in this study |
|---|---|
| Object | Active Meta ads (Facebook, Instagram, Messenger and Audience Network) that advertise residential property or property services in Marbella municipality |
| Subject | The message of each ad, meaning purpose, price disclosure, buyer filter, concrete facts, lead reason, language, location detail, call to action and destination |
| Unit of analysis | The unique creative, which groups ads from the same advertiser with the same text, because Meta lists every version of an ad separately |
| Geography | Marbella municipality, including Nueva Andalucía, Puerto Banús, San Pedro de Alcántara, the Golden Mile, Sierra Blanca, Elviria and other districts. Benahavís, Estepona and other towns are excluded |
| Period | A one-day snapshot of ads that were active on 30 September 2026 |
| Source | Meta Ad Library, all countries, all ad categories4 |
| Final sample | 430 ads, 255 unique creatives, 54 advertisers, of which 206 creatives from 43 advertisers address buyers |
The collection ran in two stages on 30 September 2026. In the first stage we ran 33 exact-phrase searches in the Ad Library in English, Spanish, German, French and Finnish together with district names, from “marbella villa” and “piso en marbella” to “nueva andalucia” and “puerto banus”, with the filters set to active ads in all countries. In the second stage every property advertiser found in the first stage was searched by name and all of its active ads were collected, which brought the total to 1,172 ads. For the few very large advertisers the collection stopped at roughly the first hundred ads in Meta's default order, so their share of the market is, if anything, understated.
Source. RealAdBook audit of the Meta Ad Library, 30 September 2026.
Each unique creative was read in its original language and coded against a codebook of eleven variables that is published with the dataset. An ad was counted as showing a price when a purchase price, a starting price or a stated buyer budget appears in the text or in the link card, and as giving concrete facts when it names at least two specifics beyond the price, such as square metres, bedrooms, the number of units, a distance or a completion date. Language, destination, start date and the EU transparency flag were taken directly from the Ad Library.
To check that the categories are stable, a random sample of 64 creatives was coded a second time, independently and without access to the first coding, and every disagreement was reviewed against the ad text. Agreement was perfect on purpose, location, price and buyer filter, between 92% and 98% on the other variables behind the main findings, and lowest on the call to action, where a bare “Learn more” button can reasonably be read as either browsing or enquiring.
| Variable | Agreement | Cohen’s kappa |
|---|---|---|
| Purpose of the ad | 100% | 1.00 |
| Property located in Marbella | 100% | 1.00 |
| Price shown | 100% | 1.00 |
| Explicit buyer filter | 100% | 1.00 |
| Two or more concrete facts | 98% | 0.97 |
| New-build | 97% | 0.93 |
| Property type | 98% | 0.98 |
| Location detail | 98% | 0.98 |
| Lead reason to buy | 92% | 0.89 |
| Call to action | 83% | 0.76 |
| Scarcity wording | 95% | 0.70 |
Note. Kappa above 0.80 is usually read as strong agreement and 0.60 to 0.80 as substantial. Both codings used the same written codebook, and disagreements were resolved by reading the ad text again.
The 206 buyer creatives come from 43 advertisers, mostly estate agencies and developers with offices on the coast together with a few portals and buyer agents. The market is concentrated, since the five most active advertisers produced 80 of the 206 creatives, or 39%. To make sure these few accounts do not decide the results, every main finding is also reported per advertiser in the results sections.
| Measure | Value |
|---|---|
| Advertisers with buyer creatives | 43 |
| Buyer creatives per advertiser, median | 3 (maximum 34) |
| Share of buyer creatives from the five most active advertisers | 39% |
| Creatives that name a district or development, not only “Marbella” | 48% |
| Creatives about new-build or off-plan property | 32% |
| Median price shown, where a price is shown | €862,500 |
| Lowest and highest price shown | €309,000 and €20,500,000 |
| Median days running on 30 September | 42.5 (middle half between 14 and 83, n = 204) |
Source. RealAdBook audit of the Meta Ad Library, active ads on 30 September 2026, n = 206 creatives, 43 advertisers.
Source. RealAdBook audit of the Meta Ad Library, active ads on 30 September 2026, n = 206.
Lowest price or budget stated in each creative.
Source. RealAdBook audit of the Meta Ad Library, active ads on 30 September 2026, n = 80 creatives with a price.
Villas dominate the advertising even though notary data for the twelve months to August 2025 show that about four in five sales in Marbella are apartments5, and the prices that are shown cluster between €500,000 and €3 million with a median of €862,500. Almost a third of creatives never say what kind of property they are about.
The first question is the simplest one, which is how much a buyer can decide from the ad alone. Fewer than half of the buyer creatives name two concrete facts about the property, and only 39% show a price. 36% give the buyer no price, no filter and fewer than two facts, which leaves a picture, an adjective and a form.
1 of the 80 creatives with an explicit buyer filter uses something other than the price, such as a minimum budget stated as a requirement.
Source. RealAdBook audit of the Meta Ad Library, active ads on 30 September 2026, n = 206 buyer creatives, 43 advertisers.
The result does not depend on a few large accounts. The average advertiser shows a price in 38% of its creatives, the share is 41% when the five most active advertisers are removed, and 21 of the 43 advertisers do not show a price in any active creative.
| Measure | Share of creatives | 95% interval | Average per advertiser | Without the top five advertisers |
|---|---|---|---|---|
| Price shown | 39% | 32 to 46% | 38% | 41% |
| Two or more concrete facts | 47% | 40 to 54% | 48% | 46% |
| None of price, filter or facts | 36% | 30 to 43% | 38% | 35% |
Source. RealAdBook audit of the Meta Ad Library, active ads on 30 September 2026, n = 206 creatives, 43 advertisers, 126 creatives without the five most active advertisers. Two advertisers tie for fifth place, and dropping the other one instead changes the facts share by one point.
Price disclosure changes sharply with the type of product. Developers and agencies selling new-build or off-plan homes put a “from” price in 58% of their creatives, while resale and single-property ads do it in 30%, and villa ads, which are the most common format in the sample, show a price in only 27%.
Source. RealAdBook audit of the Meta Ad Library, active ads on 30 September 2026, n = 206.
For every creative we recorded the one reason to act that the ad leads with. Sun, sea views, golf and the idea of a dream home lead in 58% of the creatives, and the agency's own expertise, experience or process comes second with 20%. The arguments that usually separate one buyer from another are close to absent, since investment leads in 13 creatives and relocation or a second home in 4.
Per advertiser the picture is the same, with lifestyle leading in 59% and investment in 6% of the average advertiser’s creatives.
Source. RealAdBook audit of the Meta Ad Library, active ads on 30 September 2026, n = 206 buyer creatives.
The investment ads are a small group, but they show a pattern worth noting. Only 3 of the 13 creatives that lead with investment show a price, so most of them ask the buyer to think about returns without giving the one number every return calculation starts from.
The buyer creatives are written in 14 languages. English carries 52% of them, which is expected in a market where English is the common language of foreign buyers, and the Nordic languages together carry 23%, which is more than Spanish, Dutch and German combined.
Notary data give the nationalities that actually bought in Marbella in 2025, and putting them next to the languages of the ads shows two clear gaps1. Spanish nationals made about 38% of purchases while only 10% of buyer creatives are in Spanish, and German buyers made about 7% while 2 of the 206 creatives are in German. Swedish ads sit at about half of their buyers' share and Dutch and Polish ads at about two thirds, while Norwegian ads, at 7% of creatives, are more frequent than any nationality outside the top five could account for, since the fifth-largest group made 6.67% of purchases.
Language is not nationality. Many Spanish and German buyers read English, and English ads also reach British buyers, who made 12.39% of purchases. The notary source does not state whether the nationality shares refer to all purchases or to foreign purchases only, so the comparison shows direction rather than exact gaps.
Sources. RealAdBook audit of the Meta Ad Library, n = 206 buyer creatives. Consejo General del Notariado, Portal Estadístico del Notariado, Marbella 2025, as reported by Brains RE News1.
Price disclosure also changes with the language. Spanish creatives show a price three times out of four, while English and Nordic creatives show it in roughly one case out of three. The Spanish group is small, so the interval is wide, but even its lower bound is above the English and Nordic averages.
Source. RealAdBook audit of the Meta Ad Library, active ads on 30 September 2026, n = 206.
Not every Marbella ad is looking for a buyer. Of the 255 Marbella creatives, 25 ask owners to sell or value their property, 14 offer rental or property management and 10 are brand or market content. The owner ads come from 11 advertisers and follow a recognisable script, with 19 of the 25 opening with the claim that buyers are already waiting, often with a stated budget and sometimes with a named nationality such as Dutch, Mexican or Chinese buyers.
Source. RealAdBook audit of the Meta Ad Library, active ads on 30 September 2026, n = 25 owner creatives and 206 buyer creatives.
The language split suggests two separate games on the same platform, in which listing acquisition is aimed at the local owner in Spanish while the buyer advertising is aimed abroad in English and the Nordic languages. That is an interpretation of the counts, and the data cannot show whether the buyers named in the owner ads exist.
Just over half of the buyer creatives send the person to a Meta instant form, which opens inside the app with the name and phone number already filled in, and 41% send them to a website. The share of creatives that show a price is the same in both groups at 39%, so the destination does not change how much the ad itself discloses.
Source. RealAdBook audit of the Meta Ad Library, active ads on 30 September 2026, n = 206.
Source. RealAdBook audit of the Meta Ad Library, active ads on 30 September 2026, n = 206.
The text most often asks for an enquiry, and a brochure or a price list is the ask in 11% of creatives, which in practice means that the price the ad did not show becomes the reason to leave a phone number.
The five creatives below were chosen because each one shows a pattern from the results clearly, not because they are the best or the worst ads in the market. They are quoted briefly for analysis, and each links to the full ad in the Meta Ad Library.
“Buying in Marbella or the Costa del Sol with a budget of €1M+? Tell Domus Aedes your purchase budget, preferred areas and when you want to buy.”
Asks the buyer for budget, areas and timing in the ad itself, so the wrong buyer can rule himself out before the form.
“New homes with 1 to 4 bedrooms, generous terraces and resort-style amenities, currently available from €527,500. Request current availability, floor plans and prices.”
Names the area, the number of bedrooms, a starting price and what the buyer will receive after the form.
“Et penthouse i Marbella til under €600.000 lyder næsten for godt”
A penthouse in Marbella for under €600,000 sounds almost too good to be true (Danish)
Leads with buyer representation rather than the property, and still gives a price point.
“In the heart of Marbella’s most exclusive neighbourhood, minutes from beach clubs, restaurants and Puerto Banús. Schedule your viewing today.”
A typical lifestyle creative, which names the neighbourhood and nearby places but not the type of property, its size or its price.
“No publiques tu vivienda todavía. Estamos trabajando con compradores holandeses que buscan en Marbella y alrededores”
Do not list your home yet. We are working with Dutch buyers looking in Marbella and the surrounding area (Spanish)
A listing-acquisition creative in Spanish that opens with waiting foreign buyers and a stated budget.
Interpretation by RealAdBook
The counts say what the ads contain. What follows is how we read them as practitioners who run these campaigns, and it should be read separately from the results.
When a villa ad shows the sea, the pool and the word exclusive but no price, a person with €400,000 and a person with €4 million react to it in the same way, and the difference only appears when somebody from sales calls them. In the accounts behind the RealAdBook case studies this is where most complaints about Meta lead quality begin, because the form fills look cheap in Ads Manager while the calls behind them are expensive. The market data make the point sharper, since the average Marbella sale is worth more than €700,000 and every wasted conversation is a real cost.
With 58% of creatives leading on sun, views and luxury living, an advertiser who leads with a specific reason to buy, such as the school run for a relocating family, the running costs for an investor or the airport connection for a second-home owner, stands out simply because almost nobody else is saying it. The same apartment can interest all three of those buyers, and the ad decides which of them Meta learns to find.
Developers put a “from” price in most of their ads, probably because a price list is part of how off-plan projects are sold and a starting price is a natural filter. Resale agencies rarely do, which may reflect owner confidentiality, negotiation strategy or simply habit. The data cannot tell which, but they show that price disclosure is a choice that a large part of the market already makes.
When I rebuilt the Meta account of a Marbella broker, the change that mattered most was not a new audience but an ad that told the buyer what kind of home it was, roughly what it cost and who it was meant for. Spend stayed almost the same, the cost per lead fell from $84.71 to $22.44 and the share of leads that qualified rose from about half to roughly 73%, which brought the cost per qualified buyer down from around $169 to about $31. Seeing that only 39% of the buyer ads in this audit show a price did not surprise me, because hiding the price feels safer to many agencies and produces cheaper form fills in the first week. What surprised me more is how few ads give any reason to buy beyond the view, because in a market where six buyers in ten come from abroad, a family that wants to relocate and an investor who compares yields need completely different conversations. If I were running ads in Marbella today, the first test I would set up is the same creative with and without a price range, judged by the cost per qualified buyer rather than by the cost per lead.
On 30 September 2026 the active Meta advertising for Marbella property described the place far better than the product. Most of the 206 buyer creatives led with lifestyle, 39% showed a price, and more than a third gave the buyer no price, no filter and fewer than two facts about the property. New developments were the exception, with a price in most of their ads. The advertising spoke 14 languages, but Spanish and German buyers, who together made a large share of purchases in 2025, met far fewer ads in their language than English and Nordic readers did. One Marbella ad in ten was aimed at owners rather than buyers, mostly in Spanish.
The study does not show which of these approaches produces more qualified buyers, cheaper viewings or more sales, because the Ad Library publishes none of those numbers. It also does not show what happens after the click, how the forms are built or how quickly sales teams respond. Those are the questions for the next stage, which would compare matched campaigns with and without a price, and with lifestyle against a specific reason to buy, on accounts where qualified leads and viewings are tracked in the CRM.
The Ad Library does not show spend, cost per lead or sales for commercial ads, so the study describes messaging and not results.
Ads paused before 30 September 2026 and seasonal campaigns are not included, and a longer run would show how stable these shares are.
Advertisers that never use the searched phrases in their ads can be missing, and the largest advertisers were capped at about 100 ads each.
The coding reads the ad text and the link card, so a price that appears only inside an image or a video is counted as not shown.
The five most active advertisers produced 39% of buyer creatives, which is why the main results are also reported per advertiser.
The median buyer creative had run for 42.5 days, but a long run can mean a good result or an unattended account, so running time is not used as a measure of success.
Suggested citation. RealAdBook (2026). How Marbella property is advertised on Meta, an audit of 430 active ads. RealAdBook Research, lead researcher Oleh Odukalets. https://realadbook.com/research/marbella-meta-ads-2026
Dataset and charts are licensed under CC BY 4.0 with attribution to RealAdBook. The dataset lists every ad with its Ad Library ID and link, its coded variables and the creative group it belongs to. Ad texts are not republished, and each ad can be read at its source.
This audit describes the market as a whole. Marbella Market Intelligence goes advertiser by advertiser through positioning, offers, creatives, landing pages and visible funnels, for teams that want to see how specific competitors acquire buyers.
