AI search for real estate: How France’s Orpi is driving the conversation

We know that consumers are using AI assistants to shortlist products and services online – and that the likes of Target, Wayfair, Shopify, Etsy, and Walmart have collaborated with Google to help drive such innovations. 

This shift to conversational, natural language CX is forcing brands to re-think how they bring customers into their digital ecosystems, as well as restructure their websites – and operations – to meet the needs of AI assistants. But consumers aren't just using ChatGPT, Gemini, Claude and Perplexity to buy T-shirts and laptops. Many are also using theses conversational tools to support their real estate purchases. 

French real estate cooperative network Orpi has responded by adding conversational journey capabilities to the customer experience, deployed and configured across its entire network of 1,250 agency websites. It's agentic commerce for real estate and the move makes Orpi one of the first real estate networks in the world to natively connect its catalog to AI search engines.

The conversational interface is developed by Kleio, whose technology orchestrates dozens of specialized agents connected to a unified knowledge base. The agents collaborate to analyze projects, advise users, and recommend the properties, services, and solutions best suited to each project. The overall experience is hyper-personalized, tailored and emotionally intelligent, giving house buyers a white glove service at scale for one of the most important purchases they will ever make.

This case study explains how house buyers are using AI to search for new properties, how Orpi's experience works, how the development improves the overall user experience, and the metrics Orpi is using to calculate the ROI on its conversational AI investment.

Don't miss any news, updates or insider tips from CX Network by getting them delivered to your inbox. Sign up to our newsletter and join our community of experts. 

Learn More

Machine readable digital CX for the real estate sector 

Online, machine readability is the new SEO. Without generative engine optimization (GEO) and answer engine optimization (AEO), sites will not appear in AI-first search. This is a huge blind spot. When CX Network conducted its annual research into the state of CX for 2026, AI-first customer journeys emerged as the third most influential trend for practitioners and the third most influential customer behavior.   

The new experience designed for Orpi users enables all 1,250 local agencies to showcase their catalog and territorial specificities through an assistant that unifies all of Orpi's data and historical expertise.

Orpi president, Guillaume Martinaud, says: "The real estate journey of tomorrow will be increasingly AI-assisted, and we chose to take the lead. Our clients expect a fluid search experience capable of understanding their real needs as if they were speaking to a human."

Martinaud adds: "It is a disruptive tool that leverages the immense richness of our data to better serve our clients and our 8,000 advisors, and positions Orpi where future buyers now make their first searches."

Applying conversational AI to fragmented legacy data and systems

Founded in 1966, Orpi is the definition of a legacy company and yet managed to deploy the new user experience in three months. 

As a cooperative, each agency is independently owned but benefits from shared technology, training, and branding. Not only does the new conversational UX unify data and build brand reputation, it also enables every local agency to showcase its catalog through an assistant that combines all of Orpi's data and historical expertise. For the user, it brings a standardized experience, while preserving the hyper-local touch.

To do this, Orpi partnered with Kleio create a custom AI platform. It adopted Kleio's Knowledge Engine technology layer to unify the agency network and create a seamless, friction-free customer experience. In doing so, it overcame one of the biggest challenge enterprises can face when deploying AI agents at scale: the fragmentation of data and documents across legacy systems. 

By unifying these into a hybrid search architecture purpose-built for AI reasoning, it makes complex catalogs – like a real estate network's offering – functional in agentic environments where generic AI tools fail.

The knowledge base will be progressively enriched with new data sources to meet all user needs. 

How do house hunters use AI search? 

Traditional home searches require details of the required location and price range. Additional filters may allow prospective buyers to narrow down a search according to dwelling type and additional features, such as a garage or garden.

However, consumers are now becoming accustomed to conversational search, using natural language – rather than filters – to shortlist choices. This allows for richer and more intent driven searches, and also allows buyers to escape the constraints of traditional filter-based search.

Explaining how users are prompting AI assistants, Mania Ginisty-Gold, digital and CRM manager for Orpi, says that at the simplest end of the spectrum, user prompts reflect direct intent. At the most complex end, users are looking for richer advice and insights. That means that instead of simply searching for "two to three bedrooms in Paris under one million euros", users add "close to a school and public transport", or "with strong rental-investment potential".  

Ginisty-Gold explains: "We also see a lot of questions about the properties themselves and the wider market, like price levels in a given neighborhood."

She adds: "What's striking is that most of these questions simply can't be answered by the traditional search used across real estate networks and listing portals, filters and dropdowns weren't built for them." 

In this respect, users are not simply searching for a home – they are researching a financial commitment. And as a result, because their needs are already understood and captured, sales leads reach Orpi's agents "warmer and better-qualified" Ginisty-Gold says.

"Across our other deployments, we're seeing two meaningful trends: usage keeps growing as people lean on the AI more, and conversations are getting richer, averaging seven to eight interactions each today," she adds. "Looking ahead, we're integrating simulators to handle more complex needs too, such as property price estimation or loan simulation."

Proving ROI on a conversational search experience 

The need to prove the value of AI investments has never been higher. CX Network's research into the state of CX in 2026 asked practitioners to select their three top investment priorities from a list of more than 20 choices. The most selected response was agentic AI/AI agents (selected by 29 percent), followed by automation of CX and service functions (22 percent), with data insights and analytics in fifth place (15 percent).

However, the research also found the pressure for practitioners to prove ROI is the biggest obstacle to investment. The 2026 research also found that for 52 percent of practitioners, the pressure to prove ROI is increasing.

To calculate the ROI of this investment, Orpi isn't just looking at CX, financial and business metrics, such as net revenue retention (NRR), net promoter score (NPS), or customer retention. 

Ginisty-Gold says the main focus will be on "lead quality and conversion, engagement, and agent productivity, ultimately, how many conversations turn into viewings and [instructions]". 

She adds: "There's also a forward-looking angle: the goal is to connect Orpi's AI agents to the AI agents of public search engines like ChatGPT and Google AI (agent to agent), where customers increasingly start their search. Because Orpi's agents are trained on its own catalog and customers, they will be more relevant and will surface better there, so part of the ROI will come from new traffic captured upstream."

Quick links

• How generative AI search is changing the customer journey
• How do you rank in ChatGPT, Claude, and Perplexity?
• 5 brands transforming CX and service with conversational AI