Average CPC on 'real estate estimate' keywords in France Q4 2025: between €3.80 and €7.20 depending on tight zones, with a median lead → signed mandate transformation rate observed between 8 and 14% on accounts we track (2025-2026 aggregated Google Ads data). On buyer keywords ('apartment for sale [city]'), CPC drops to €0.80-2.40 but the lead → compromis transformation rate doesn't exceed 2 to 4% — a buyer consults 4 to 7 agencies before signing. This asymmetry changes everything for steering a 2026 real estate Google Ads account
French real estate emerges from 24 months of crisis (ECB key rates, transactional freeze on existing, 18 to 22% drop in sales volume 2023-2024 per notaires de France). 2026 marks recovery, but with modified demand structure — sellers are more reluctant to undersell, buyers more demanding on qualification, cycles lengthen. For agencies, developers and marketplaces, Google Ads becomes again a contestable acquisition channel but with new requirement: steer on real commission, not on raw lead. This article rolls out the complete method, from intent choice to offline tracking up to ROAS calibration on long cycle. For measurement fundamentals, see our CRM offline conversions guide as a complement
Real estate in 2026: SEA context after the crisis
French real estate in 2026 is a 2-speed market. New construction remains constrained by the progressive end of the Pinel scheme, rising construction costs (+22% cumulative 2021-2025) and developer inertia on programs launched before the crisis. Existing recovers faster, supported by stabilization of ECB key rates around 2.75 to 3.25% per the Banque de France in Q1 2026, but with transaction volume remaining 12 to 18% below the 2021 peak. For Google Ads, this means the return of competition on seller bids after 2 years of disengagement by many agencies.
Concretely, CPCs on seller intent keywords ('free home estimate', 'how much is my apartment worth', 'sell quickly [city]') rose 35 to 55% between Q4 2024 and Q1 2026 across zones we track. CPCs on buyer keywords stagnated or even slightly declined, because marketplaces (SeLoger, Leboncoin, Bien'ici, Logic-Immo) absorb a growing share of buyer intent via SEO and vertical portals. For an independent agency, the consequence is clear: it has become more profitable to chase sellers (where marketplaces have less structural advantage) than buyers (where they dominate).
The 3 business segments and their 2026 SEA realities:
- Independent real estate agencies — commission-per-mandate model, ICP residence owners in mobility (move, succession, divorce, larger purchase). Target seller CPL: €80-160, transformation 8-14%, average mandate value €5-12k.
- Franchised agency networks (Century 21, Laforêt, Orpi, Guy Hoquet, etc.) — head office centralized budgets + local budgets per agency. Often in Google Ads MCC to steer dozens of local accounts simultaneously.
- New construction developers — program sale model, ICP first-time buyers or LMNP investors. Target buyer CPL: €40-90, transformation 2-5% on showroom appointment, sold lot value €200-450k.
- Marketplaces and portals — model selling leads to agencies or display ads. Audience-first logic, not relevant for this article focusing on direct advertisers.
For independent agencies, it's the unit economics gap (LTV €5-12k vs CPC €4-7) that makes Google Ads structurally attractive — provided you steer on real commission and not on lead volume. This discipline is precisely what most public benchmarks miss.
Buyer intent vs seller intent: 2 opposite strategies
The most frequent mistake in real estate Google Ads: treating buyer intent and seller intent as a single market. They are two opposite markets in CPC, CR, LTV, geographic targeting and ad creative. A campaign mixing both in the same ad group is a campaign underperforming on both fronts simultaneously — Smart Bidding can't optimize two contradictory objectives.
Seller intent (estimate, mandate) — the gold mine:
- Typical keywords: 'house estimate [city]', 'sell apartment quickly', 'how much is my house worth', 'real estate agency free estimate', 'exclusive vs simple mandate'.
- Typical monthly volume in mid-size zone (50-150k inhabitants): 800 to 2,500 cumulative monthly searches.
- Median CPC: €3.80 to €7.20, with peak at €9-12 on hyper-intent queries ('sell house urgent [city]').
- Landing conversion rate: 6 to 12% depending on form quality and incentive (free estimate report, expert appointment).
- Lead → signed mandate transformation: 8 to 14% over 30-60 days.
- LTV: average commission €5,000 to €12,000 excl. VAT, plus volume 2025-2026 stabilized.
Buyer intent (alert, viewing) — volume but little captivity:
- Typical keywords: '3-bedroom apartment for sale [neighborhood]', 'house with garden [city]', 'real estate [city] cheap', 'properties for sale [postal code]'.
- Monthly volume: massive, but largely captured by SeLoger/Leboncoin/Bien'ici/Logic-Immo.
- Median CPC: €0.80 to €2.40, much more accessible.
- Landing conversion rate: 3 to 6% (the buyer wants to compare, not commit immediately).
- Lead → compromis with your specific agency transformation: 2 to 4%, because they consult 4 to 7 agencies in parallel.
- LTV: similar €5-10k commission but probability that YOU are the final agency, low.
At equivalent budget, $1 spent on seller intent generates 3 to 5x more cashed commission than $1 spent on buyer intent — even when seller CPC is 3 times higher. The reason: captivity. A seller signing an exclusive mandate with you is captive until sale. A buyer remains a butterfly between 4 to 7 agencies. Favor the seller in 60-70% of budget for a classic agency, unless you have a structural advantage on acquisition (e.g. unique exclusive portfolio in region).
Structural implications:
- Separate campaigns by intent. No mixing. 'Seller' Search and 'Buyer' Search are 2 distinct Google Ads campaigns, with distinct budgets, distinct creatives, distinct CPA targets.
- Separate landing pages. A 'Free estimate in 48h' landing for sellers, a 'Find your future home in [city]' landing for buyers. No common page. See our complete landing pages conversion guide.
- Separate Smart Bidding. Different Target CPA: €80-160 on Seller, €25-50 on Buyer. The same CPA target would be absurd on both.
- Different audience targets. For Seller, confirmed property owners (in-market 'Real Estate / Sellers' audiences). For Buyer, in-market 'Real Estate / Buyers'. Different Customer Match strategy on each flow.
For new developers, the logic reverses: 80% Buyer (you have a program to sell, no need for mandates), 20% notoriety/brand. Classic agencies reverse: 60-70% Seller, 30-40% Buyer.
Hyper-local geo-targeting: city, neighborhood, radius
Real estate is essentially a hyper-local business. An agency in Lyon 6th arrondissement has no chance of serving a mandate in the 9th — the real sales zone is 2 to 8 km maximum in dense urban zone, 15 to 30 km in peri-urban zone. For Google Ads, this means standard 'France' or 'Auvergne-Rhône-Alpes' geo-targeting is a budget catastrophe. Geo precision is as important as keyword choice.
The 3 levels of Google Ads geo-targeting for real estate:
- City targeting (INSEE municipality) — the simplest, automatic exclusion of irrelevant neighboring municipalities. Relevant for mid-size cities (30 to 200k inhabitants) where sales zone = municipality.
- Radius targeting — around the agency address, typically 3-5 km in dense urban, 10-20 km in peri-urban. More flexible than municipality targeting, especially at administrative borders.
- Neighborhood or postal code targeting — for Paris, Lyon, Marseille where the municipality is too large. Target by arrondissement or postal code cluster. Hektor and Apimo often allow exporting requests by micro-zone for calibration.
Geo calibration procedure in 4 steps:
- Identify real sales zone from your CRM. Extract the 100 most recent signed mandates and geolocate. The zone containing 80% of mandates = your Google Ads zone.
- Define corresponding radius or municipality list. Favor explicit list if zone isn't circular (e.g. exclusion of an irrelevant neighborhood).
- Activate 'People in targeted location' option (not 'People interested in location'). Otherwise you reach also returning expats searching to buy in your city but residing 800 km away — relevant for Buyer, not for Seller.
- Bid modifier per zone on Search Buyer. Push +20-30% on postal codes where your stock is concentrated, lower -15% on secondary zones.
Targeting 'whole France' on Search Seller thinking to scale. Catastrophic result observed in audit: 75 to 85% of budget wasted on queries outside sales zone (a Bordeaux owner asking for an estimate from your Paris agency will never become a mandate). Strict geo filtering is non-negotiable in real estate. If you want to scale geographically, create 1 campaign per zone, not 1 France campaign.
Particular case of franchised networks: a network steering 80 local agencies must use an MCC account with 1 sub-account per agency (or per agency cluster) to allow distinct geo steering, variable local budgets, and proper lead attribution to the concerned agency. For complete MCC strategy, see our multi-account MCC strategy guide.
Google Ads lead forms vs landing page: quantitative comparison
The native Google Ads lead form is a form that opens directly in the SERP or an ad extension, without redirection to the site. The user clicks, sees a form pre-filled with their Google info (name, email, phone), validates in 2 clicks. Official documentation on Google Ads support. It's the ideal tool for top-funnel volume. But in real estate, its use must be nuanced.
Lead Form vs Landing Page real estate performance comparison:
Reading the table: the lead form generates 2 to 4x more raw leads at 50% lower CPL — magnificent on the surface. But the qualification rate collapses: half of leads aren't reachable (wrong phones, fake numbers, junk emails), and among reachables, the transformation rate falls because the contact didn't express real effort. Ultimately, the mandate CAC is roughly the same between the two devices — but with 2 to 3x more commercial work to handle the Lead Form junk volume.
Our recommendation by use case:
- Google Ads lead form — for broad information keywords ('price per sqm [city]', 'real estate market [city]'), for new developers wanting to fill a prospect file to recontact over 6-18 months, or for agencies wanting to test a new geo sector without investing in a landing page.
- Dedicated landing page — for high-intent transactional keywords ('sell house quickly', 'urgent apartment estimate'), for any serious mandate acquisition campaign, for premium agencies filtering by effort. Remains the standard.
- Hybrid — main landing page + lead form in ad extension on certain keywords. Allows capturing both profiles on the same campaign.
Building the converting landing page:
- Clear promise in H1 ('Free no-obligation estimate of your property in 48h').
- 1 single form above the fold, 4-5 fields maximum (name, phone, email, property address, type).
- Reassurance element: team photo, years of experience, number of mandates signed in 2025.
- 1 recent seller testimonial with first name + neighborhood.
- Explicit call to action: 'Request my estimate' (not 'Send').
- Locality proofs: sector map, neighborhood mention in H2.
- Clear GDPR compliance with consent checkbox.
To go further on landing mechanics, see our complete landing pages conversion guide. Landing quality determines 60 to 70% of final CAC in most public benchmarks.
CRM qualification: Leeloo, Apimo, Hektor integration
The CRM is the most important qualification tool for a real estate Google Ads account. Without it, you steer on raw lead — which amounts to optimizing on noise. The clean lead → mandate → compromis → deed chain requires native integration between Google Ads, the web form and the trade CRM. In France, 4 to 5 tools dominate the agency market and 2-3 tools dominate the developer market.
Major French real estate CRMs in 2026 and their Google Ads compatibility:
- Hektor (La Boîte Immo) — independent agencies and small networks market share leader. Native GCLID custom property, clean CSV export, increasingly mature API. Compatible with offline conversion upload without custom dev.
- Apimo (Immo-facile) — mid-market agencies and networks. Configurable custom fields, timestamped exports. Requires some setup on the web form side to transmit GCLID via webhook.
- Leeloo (HabitatPresto) — gaining momentum 2024-2026. Native integrations with Zapier, Make, n8n. Suited for agencies that don't want to develop.
- Netty (LeMoniteur) — new construction developers, complete program management. Data segmented by program/lot, perfect for steering CAC per program.
- Périclès (agency network) — reserved for franchisees having it as standard. Consolidated multi-agency reporting.
- Adapt'immo, Immopros, Krea, La Boîte Immo Premium — less widespread alternatives but with correct API integrations.
Generic integration procedure (valid for the 5 majors):
- Create a custom Contact or Prospect property named
gclid(free text, 100 characters). - Add to web form a hidden
gclidfield populated via JavaScript from the stored Google Ads cookie. - Configure the form's webhook or API push to the CRM with
formGCLID → CRM.gclidmapping. - Test with a real Google Ads click + form submission + CRM verification (the GCLID must appear in the prospect file).
- Define CRM statuses corresponding to conversions to upload: Qualified Lead, Scheduled Appointment, Signed Mandate, Signed Compromis, Final Deed.
The critical rule: the GCLID must follow the contact throughout the cycle. If the contact moves from prospect to qualified lead in the CRM, the GCLID is preserved. If the lead becomes a mandate, the GCLID is preserved. Otherwise, the offline conversion upload won't be able to surface to the original click — and all tracking is broken.
Hektor, Apimo and Netty automatically dedupe prospect files by phone or email. If the same contact fills 2 different forms (1 seller + 1 buyer), files merge — and only one GCLID is preserved. Configure your CRM to store a GCLID history (multi-values) or create a file per form before manual merge. Otherwise, you lose 15 to 25% of attributions on contacts doing both processes.
For advertisers wanting an in-depth guide on CRM integration on the tracking side, see our CRM offline conversions Google Ads guide. The methodology is identical for B2B SaaS and real estate — only fields and values change.
Offline closing tracking: from lead to compromis
Offline tracking in real estate follows the same principle as in B2B SaaS: surface the final conversion (compromis or mandate) up to the original click via GCLID, and upload this conversion to Google Ads to calibrate Smart Bidding. But real estate adds 2 specifics: longer conversion window (90-180 days), and plurality of conversions to measure (appointment, mandate, compromis, deed).
The 4 offline conversions to upload, with value:
Optimal upload method:
- Weekly — CSV CRM export every Friday evening: all status changes of the week (new appointment, new mandate, etc.). No need for daily — Google Ads aggregates over the extended conversion window.
- Required CSV format — columns: Google Click ID (GCLID), Conversion Name, Conversion Time (date+time ISO 8601), Conversion Value (in euros), Conversion Currency (EUR).
- Upload via UI or API — Google Ads UI (Tools > Conversions > Import) is enough for accounts with under 5,000 conversions/month. API is necessary beyond or to automate via a Python script or Make/Zapier.
- Verification — weekly verify that conversions surface in Google Ads (Conversions > Imports). Max appearance delay: 24-48h.
Extended conversion window configuration:
By default, Google Ads attributes conversions over 90 days after click. In real estate, that's insufficient — a compromis is sometimes signed 120-150 days after the first search. The window must be extended to 180 days minimum, parameterizable at conversion level (Conversions > Details > Conversion window). Without this, you lose 18 to 28% of attributions on long cycles.
Smart Bidding calibration on final conversion:
Once offline tracking is mature (30+ Mandate or Compromis conversions surfaced over 90 consecutive days), switch Smart Bidding from Maximize Conversions (or Manual CPC at startup) to Target CPA calibrated on historical mandate CAC. For an independent French agency, the median observed mandate CAC in public benchmarks sits between €800 and €2,000 depending on zone and property segment (entry-level vs premium). The classic trap: calibrating Target CPA on CPL and not on mandate CAC. Invariable result: the algorithm finds inexpensive junk lead that never transforms, real CAC explodes after 60 days.
For technical conversion tracking fundamentals, see our Google Ads conversion tracking guide.
Real estate ROAS: how to measure a sector with 6-18 month cycle
The ROAS question in real estate divides beginner advertisers. Many steer on apparent CPL ('my CPL is €50, I'm happy'), which is a structural mistake. Real ROAS measures on cashed commission, not on generated lead. And commission is only cashed 6 to 18 months after the Google Ads click, requiring patient ROAS modeling.
The 3 ROAS horizons to model:
- 30-day ROAS (lead→appointment) — purely operational, validates that tracking and the top funnel work. Target: 1 appointment per €8-15 of spend.
- 90-day ROAS (lead→mandate or compromis) — that's the ROAS Smart Bidding can steer with the extended conversion window. Target: 4-6x on expected commission (commission / Google Ads spend).
- 12-month ROAS (lead→cashed commission including bounces) — the real business ROAS, accounts for seller leads bringing a mandate 6 months later, buyer leads finalizing a compromis on the next shown property. Target: 8-15x on mature accounts.
Practical modeling: the 30/90/12-month matrix.
The operational dashboard of an agency seriously steering Google Ads has 3 ROAS columns for each campaign:
- 30-day apparent ROAS (useful to validate tracking and top funnel health).
- 90-day weighted ROAS (used to steer Smart Bidding via Target CPA mandate).
- 12-month final ROAS (used for quarterly budget arbitrations and the decision to scale).
Complementary indicators to track:
- CR mandate → effective sale — not all agencies sign all their mandates. Rate varies from 55 to 78% depending on portfolio quality and market. Multiply Google Ads CR by this factor to obtain real final ROAS.
- Median delay mandate → sale — in 2026 on the FR market, median 90-150 days depending on zone and segment. To integrate into the extended conversion window.
- Positive bounce rate — seller clients who become buyers (move) or vice versa. Adds 12 to 22% cumulative LTV on mature accounts.
To steer this level of finesse, launch a free SteerAds audit that specifically evaluates offline tracking maturity and Smart Bidding calibration quality on your real estate vertical. Most public benchmarks have a 25 to 45% improvement margin on 12-month ROAS simply by correcting the tracking chain and target calibration.
For ROAS fundamentals, see our ROAS CPA CPC guide. For very high-intent premium keywords ('sell house urgent') where Quality Score conditions acquisition cost, see our Quality Score guide. And for agencies still hesitating with the local services angle, the parallel with law firms is enlightening — same long cycle logic on local intent.
Real estate 2026 is a sector where Google Ads can become massively profitable again, provided you accept it's a patience sport. Six months to validate a campaign, twelve months to adjust allocation, eighteen months to scale cleanly. Agencies steering on cashed commission rather than raw lead will gain 3 to 5 years of advance over those staying at apparent CPL.
Sources
Official sources consulted for this guide:
FAQ
Should you favor seller leads or buyer leads in 2026?
Both, but with separate budgets and KPIs. The seller lead (free estimate, listing mandate) remains 3 to 5x more profitable in LTV for an agency — average commission per mandate runs between €4,000 and €12,000 excl. VAT in the provinces and €8,000 to €25,000 in Île-de-France. The buyer lead (alert, viewing) is more volume-based but less captive: a buyer consults on average 4 to 7 agencies before signing a sales agreement. Our median observed allocation on French mid-market agencies: 55-65% of Google Ads budget on seller keywords ('apartment estimate', 'how much is my house worth') and 30-40% on geographically targeted buyer keywords. New construction developers reverse the logic: 80% buyer because they don't need mandates.
Google Ads lead form or dedicated landing page for real estate?
Lead form for top-funnel volume, landing page for bottom-funnel qualified. The native Google Ads lead form typically generates 2 to 4x more raw leads at equivalent CPL, but the lead → signed mandate transformation rate drops 40 to 60% compared to a landing page. Why: the pre-filled lead form captures lukewarm intentions (a user who just clicked without thinking) while a landing page filters by effort. Our method for agencies: lead form on information keywords ('price per sqm [city]'), landing page on transactional keywords ('sell house [city] urgent'). For new developers with showroom, systematic landing page — form quality conditions the physical appointment.
How to measure Google Ads ROAS on a 6 to 18 month cycle in real estate?
By surfacing offline conversions from CRM up to the original click via GCLID. The clean chain: Google Ads auto-tagging activated, GCLID capture in the form, storage in dedicated Contact property in Leeloo, Apimo or Hektor, then weekly upload of signed deals (compromis, mandate, sale) to Google Ads via the offline conversion interface or API. The Google Ads lookback window is 90 days by default — that's insufficient for real estate. Activate the extended window to 180 days minimum, and steer Smart Bidding on the signed compromis conversion, not on the raw lead. Without this setup, you optimize on junk leads and the real CAC is 3 to 5x higher than the displayed CPL.
What minimum monthly budget for Google Ads in a real estate agency 2026?
€1,200 to €2,500/month minimum for a serious 90-day test on a single sales zone (1 mid-size city or 2-3 dense urban neighborhoods). Below this threshold, conversion volume remains insufficient to steer Smart Bidding on competitive keywords ('real estate agency [city]' is regularly at €4-8 CPC in tight zones). Above €5,000/month, the priority becomes offline tracking maturation and diversification of long-tail keywords by micro-neighborhood. Agencies scaling well in 2026 generally exceed €8,000-15,000/month on 1 to 3 zones and capture 12 to 20% of mandates in their sector via digital.
Is Performance Max relevant for a real estate agency?
With caution and only after maturing Search. PMax is tempting for scaling top-funnel (estimate, alert) but it poses 2 problems in real estate: brand Search cannibalization is high (users often type the agency name after seeing a physical storefront), and the real estate product feed isn't standardized in Google Merchant — most agencies have nothing to give PMax in structured signals. Our method: no PMax before 90 days of matured Microsoft + Google Search, mandatory brand exclusions from activation, and strict cap at 25-30% of total budget. Beyond, PMax buys back leads already acquired via local SEO and Google Business Profile.