Verticals under AI Overviews: the 2026 real estate playbook for e-commerce
Summarize this article with AI
A client calls me: $8,000 invested in guides, and yet the click disappears
A client calls me on a Tuesday morning. He put $8,000 into e-commerce guides. Comparisons, tests, so-called « store » content. His pages rank. Traffic rises slowly. Then AI Overviews arrive. The answer is generated. The click disappears.
He tells me: « Stéphane, I did what everyone advises. »
I answer him: « Your problem isn’t the content. It’s the architecture. »
Neil Patel says it in his 2026 Real Estate Marketing guide. Zillow, Realtor.com, Redfin take almost all high-volume queries. AI Overviews crush what was left of informational clicks.
Look at real estate. It’s the perfect lab.
Three marketplaces. Well-structured listings. Prices, photos, reviews, a location. Google no longer needs to send the user to a blog that explains « how to buy an apartment in Lyon. » It cites the listing directly. It compares. It ranks.
The click goes to the transaction.
This mechanism isn’t only about property. It concerns your catalog. Every product page is a miniature real estate listing. It has a price, availability, a rating, a logical place in the catalog. It attracts the click when Google can cite it in AI Overviews. Otherwise, it stays invisible under the answer.
For an e-commerce client, the observation was clear. 47 transactional queries captured by product pages. Before the semantic restructuring, those queries went to generic blog posts. 4,200 organic sessions per month. After the restructuring, 11,800 sessions. Observed order of magnitude.
The real estate playbook doesn’t say « produce more. » It says « structure better. »
That’s it.
The real estate lab: Zillow, Redfin, Realtor and click compression
Real estate is the vertical that shows the future of all the others. Why? Marketplaces have held the transactional data there for twenty years. Zillow, Realtor.com, Redfin have structured every listing. Price, photo, surface, location, history, value projection. Neil Patel says it very simply in his 2026 guide: these players occupy almost all high-volume queries.
When Google launches AI Overviews, it doesn’t need to explore a thousand sites. It has three stable sources. It aggregates. It answers. The informational click melts.
The lesson goes further than « marketplaces win. » They win because they provide the data the algorithm can cite without checking a blog page.
An apartment in Lyon. Three rooms. €340,000. Croix-Rousse district. A 4.7 rating. The AI Overviews summary contains exactly these elements. The site that wants to capture this query must have the same structure. Otherwise, it blends into the landscape and the click goes elsewhere.
For an e-commerce client, I see the same mechanic. A catalog of 800 references. Product pages with a price, a stock level, a review. Yet no semantic structure links these pages together. Each product stays isolated. Google cannot consider it a stable entity within a set.
Semantic restructuring answers that. We link each product page to its category, its uses, its alternatives. We enrich structured fields. We create an internal linking that signals to Google the place of each page in the whole. The real estate pattern becomes reproducible.
Traffic comes back through architecture.
The invisible mechanism: AI Overviews don’t steal, they reorganize intent
I often hear: « AI Overviews are stealing our traffic. »
False.
They compress informational intent. Transactional traffic remains. It moves toward pages that hold structured data.
In real estate, a query « buy apartment Lyon 3 rooms » triggers a local response. A map. Three listings. A price comparison. The blog that explained « how to buy in Lyon » no longer appears. It isn’t a penalty: it lacks the data the algorithm cites.
Neil Patel’s 2026 playbook shows it. Marketplaces win because they provide three transactional signals before the click: price, availability, location.
Three signals. Not seven. Not eighty.
The rest of the peripheral content qualifies. The capture happens elsewhere.
Same mechanic on an e-commerce product page. Price. Stock. Rating. Lead time. Compatibility.
When your product page has these structured fields, it becomes the source Google cites in AI Overviews. It attracts the transactional click.
Content has nothing to do with it. Architecture decides.
I build semantic cocoons around product data. Each pillar page captures purchase intent. Each supporting page feeds the pillar with context.
For one of my e-commerce clients, the pillar product page captured 2,800 clicks per month after restructuring. Before, it was buried in pagination. No internal link supported it. The difference is rarely in the words. It’s in the internal linking.
Un catalogue sans structure sémantique, c’est un quartier sans plan : les pages existent, mais rien ne les relie. Voici l’architecture qui tient : un pilier par intention, des satellites qui font le lien.
Le cocon sémantique : pilier et satellites
Un pillar, vingt clusters. Les trois constellations dorées captent 42 % des citations AI Overview. Survolez pour explorer.
The winning structure: three transactional signals before content
In real estate, a listing that performs doesn’t start with a long text. It starts with a price, a photo, a neighborhood, a publication date. The narrative comes later, to reassure or to stand out. Transactional data comes first.
In e-commerce, it works the same way. A product page first answers three questions: Is the price right? Is the product available? Is it compatible with my need? AI Overviews pull these elements from structured fields. And they display them first.
A catalog of 800 references without semantic structure is a neighborhood without a map. The pages exist. No one links them. Google doesn’t understand the hierarchy of intent. The click goes to a competitor with an enriched product page, linked to its category page, itself linked to a usage guide.
The real estate playbook distinguishes three levels of pages:
- The listing: price, surface, neighborhood, photos.
- The market page: price per square meter, trends, lead times.
- The informational guide: steps, advice, tax.
Under AI Overviews, what captures is the listing. It answers the main intent. In e-commerce, the product page plays that role. It must be linked to a market page and a usage guide. Semantic internal linking guides crawl and ranking.
A product page without a link to its category is a real estate listing without a neighborhood. It loses context. It loses the data the AI exploits.
DOSE transposition: extracting the real estate pattern for e-commerce
The DOSE framework, learned with Guillaume Attias at BMO Academy, starts from a simple idea. A brain retains an abstract pattern better when it meets it in another vertical. Zillow isn’t copied. Its mechanism is extracted.
Why real estate for e-commerce?
A transactional vertical held by marketplaces. E-commerce lives the same thing facing marketplaces. And in most niches, Google can assemble the answer itself.
The transposition comes down to three rules.
- Observe the pages that capture under AI Overviews in real estate.
- Identify the signals they aggregate: price, availability, reviews.
- Reproduce this architecture on your product pages.
Nothing copy-pasted. A work of abstraction.
For a furniture seller client, we transposed the pattern. The product page became a listing. Stable price, visible stock, aggregated rating, description linked to usage rooms. The category page took the role of market page. Decoration guides moved to the periphery.
The result settled in stages. Nothing brutal. And it holds.
The vertical analogy avoids waiting for data from your own vertical. You move before the pattern jumps out at your competitors.
Here, no method to sell: I show you the pages.
The 2026 playbook in 4 applicable moves
How to transpose the real estate playbook to e-commerce, concretely.
First move: map the transactional queries under AI Overviews.
Look for the 47 queries that trigger a product response. 1,000 keywords are useless. Search Console provides this data. Filter impressions whose average position drops below 5. That’s where informational compression begins.
Next: restructure product pages into semantic cocoons.
One pillar page per purchase intent. Satellite pages for context. Internal linking that connects each product to its category and its use.
Then tag the structured data the AI cites.
Price, availability, ratings, compatibility, lead time. These fields are mandatory. They are the raw material of AI Overviews. Google cites them, then users click.
Last point: measure the evolution by click type.
Look at transactional traffic. Product pages. Category pages. Organic conversions, fully organic. The overall volume tells less.
A client observed: « My total traffic dropped 12%. My organic conversions increased 38%. »
That’s the signal.
The 2026 real estate playbook refocuses visibility on the transaction. It lets informational traffic go.
Structure before your competitors. That’s where you win.
Un avant/après vaut mieux qu’un long discours. Voilà ce que change concrètement une refonte sémantique sur le trafic organique d’un client : tout vient du référencement naturel, sans levier payant.
De 4 200 à 11 800 sessions organiques
L’effet mesuré d’une refonte sémantique sur le trafic d’un client e-commerce.
E-commerce visibility plays out on data, not text
I often hear: « Real estate is different. Every property is unique. E-commerce is a standardized catalog. »
I look at the listings. They look alike.
A property has a price, a photo, a neighborhood, a listing date, an agent. An e-commerce product has a price, a photo, a category, a stock date, a seller. Same structure. The scale differs.
Google looks for a stable entity to cite. A well-structured product page is one. A real estate listing is another.
The 2026 real estate playbook taught me one point for e-commerce: transactional visibility comes from data.
My clients who restructured their product pages before the others take the clicks in a less competitive market. Their advantage comes from the product page, denser. The assistant finds material to cite there.
Before publishing your next guide, check the associated product page. Is it rich enough to deserve the click?
And what if your next semantic restructuring started with the product page, rather than the blog post?
An audit of your product pages for AI Overviews
I look at your 47 highest-value queries and show you which product pages to restructure first to capture transactional clicks.
Book a strategic call — 45 minFrequently Asked Questions
Why does real estate serve as a laboratory for e-commerce under AI Overviews?
Real estate is a transaction market held by marketplaces. They have been exploiting data for years. AI Overviews cut informational clicks there first. The same pattern transfers to e-commerce product pages.
Which transactional signals should I surface on my product pages?
Price, availability, rating, lead time, compatibility. I observe the same thing every time: these are the fields Google picks up most in AI Overviews, just before sending the click.
Should you stop producing guides and blog posts?
No. You need to reposition them. Guides play a peripheral qualification role, while the product page becomes the pillar page that captures transactional intent.
What is a semantic cocoon applied to a product catalog?
A pillar product page that captures purchase intent. Around it, satellite pages for context: category, usage, alternative, guide. And internal linking between all of it.
How do I measure the impact of this restructuring beyond overall traffic?
Look at transactional traffic: product pages, category pages, conversions from organic only. Informational traffic drops and conversions rise? That’s the right signal.
