Of all the industries facing AI-powered disruption, real estate brokerage might be among the ripest.
There was a time in the not-too-distant past when people looking to buy or rent homes would start by calling realtors. Not anymore. Today, clients are far more likely to begin their buying journeys online, where richly populated sites enable rapid, broad property searches and custom filtering across dozens of desired features and attributes. In addition, potential buyers or renters can browse properties in their own time and even take virtual house tours.
And, while many sites will eventually loop in human agents, the disintermediation risk for realtors — especially small and mid-size agencies — is real and rising
The fundamental shift to online browsing for properties — multiplied across time and space — is generating vast quantities of data. For example, in the U.S. alone, the Top 4 consumer real estate sites — Zillow, Trulia, Yahoo!, and Realtor — together average nearly 100 million unique visits per month. Imagine if every site visitor looks at ten properties, on an average, and can filter searches based on 30-40 different property attributes. Very quickly, you’re into billions of raw data points which contain a wealth of information around the preferences and intentions of buyers and sellers in the marketplace
You can be sure the tech-powered disruptors are exploiting AI in real estate to tap the deal-making insights locked in that raw consumer data. While traditional realtors may still have the edge in people and client relationships, to survive and thrive long-term with an edge over margins, they will need to become much more:
Operationally efficient and automated
Focused on finding new ways to differentiate the buying experiences they create for clients
Technology — featuring AI — is a common denominator for moving forward on all three of these imperatives. On the agent side, a tool such as a real estate CRM will typically address many basic automation and efficiency aspects, but agents still need to be using it consistently, as intended. AI in real estate, meanwhile, combines both sales efficiency and data-powered intelligence, by:
Assisting agents with prioritizing deals most likely to close
Suggesting specific actions that are most likely to advance deals from one pipeline stage to the next
Alerting agents as to when they should begin certain processes (such as preparing deal paperwork) to ensure it’s ready when needed
On the customer journey side of the equation, there is also enormous potential for AI in real estate to improve experiences. A few examples:
Extending home search capability beyond what can be searched manually
Dynamic surfacing of options such as ‘new similar property now available’ based on what a buyer has searched or appreciated in viewing other properties
Analyzing and pulling additional information from property images that would not be included in listings — for example, painted versus unpainted woodwork, style of lighting fixtures, estimated ages of appliances, difficult-to-furnish spaces, quality of landscaping/property maintenance
The good news for people-powered realty agencies is that real estate CRM systems are increasingly incorporating the same kinds of AI technologies — machine learning, natural language processing, chatbots — that the digital industry disruptors are exploiting. Here is a quick review of eight high-impact opportunities to drive value for both agents and buyers through a typical home-sale or buy process.
Agents often struggle to filter and prioritize leads that originate from multiple sources such as websites, digital advertising, landing pages, or paid property-search sites and other external sources.
A basic real estate CRM enables lead scoring based on a series of actions that a prospect takes. For example, you might assign points for click-throughs from promotional emails or web ads for new listings, mortgage advice, open house invites, filling out web forms, engaging with chatbots, and so forth. But setting the criteria for that model is still arbitrary, as it draws more from human experience than journey-driven insights about when a name in a database converts to a sales prospect.
Now, imagine a CRM, unifying marketing and sales functions, that captures every single interaction across a buyer’s entire journey, extending even beyond deal closing. Add an ML algorithm that constantly studies all that buyer journey data for subtle patterns and connections between buyer actions and deal outcomes that deliver substantial profits to your agency.
Instead of arbitrarily deciding how to score incoming leads, you’re now taking a data-driven approach to determining which leads are most likely to result in favorable deals, while giving agents more time to focus on closing deals.
Showing properties consumes time, and — as the adage goes — time is money. And, while there may be some strategy to showing a few less-than-perfect properties — leading up to the perfect one — there are enormous operational efficiencies and productivity gains to be found in using AI in real estate to assist agents with fine tuning the selection of properties.
Like the algorithms that suggest movies or shows to watch on Netflix or companion products to buy on Amazon, AI in a real estate CRM can assist with finding and matching properties to clients based on expressed preferences, demographic factors, and observed behaviors (such as viewing and rejecting a specific property for X, Y, or Z reason). With personalized AI recommendations, agents can focus on showing only the most relevant properties, which helps in buyers staying engaged for longer.
The toughest part of a real estate-buying journey is deciding which property options to pursue and how aggressively to bid. Loving a property and envisioning themselves living there is not enough; most buyers would also want to be assured they’re making a sound, long-term investment. Such an assurance moves them toward making an offer.
Digital property listings follow standard templates, detailing all sorts of common things. But those listings often omit many qualitative parameters that make homes desirable over the long term. For example:
Access to public transportation (especially in suburbs)
Availability of parking, public parks (especially in cities)
Local dining places, recreation, nightlife/entertainment, good-quality schools
Potential for future development of surrounding properties
Optimal financing options
By combining CRM data with marketplace data — for instance, transportation systems, area crime statistics, reviews and performance stats for local schools, and local marketplace activity — AI in a real estate CRM can help agents and brokers to better predict future home values and to analyze properties in depth, leading to more deals, positive reviews, and referral business for your agents and agency.
Beyond learning more about specific properties, advanced property analysis needs to be predictive, for example, anticipating future rent or sale price fluctuations in specific markets. Four other data-driven pursuits ripe for AI in real estate include:
Pricing properties for sale or rent
Setting successful bid strategies for desired properties
Deciding when to list/de-list properties
Predicting ROI (or lack thereof) on property investments
Imagine a scenario of three parties embroiled in a bidding war due to very tight home inventory. One of the parties has an agent using an AI-powered real estate CRM, which in turn, uses data about the property and local market conditions, to suggest optimal bid increments and timing, while the other two rely only on agent experience and gut feel. All three parties might love the property, but only one is going to know for sure when it no longer makes sense to continue bidding up the price. And should that party decide to override the rational choice, they would still be doing so explicitly with AI-backed data versus being swept along by emotions.
Another big focus area for AI in real estate is on pipeline management and forecasting. For example, ML algorithms can analyze sales pipeline activity to:
Flag deals with rising risks of turning cold based on factors such as similar past deals, local area trends, and past engagements with the buyer
Auto-generate recommended actions that are most likely to succeed at re-warming deals to help with pipeline progression
This benefits the agency by improving revenue forecasting accuracy. For agents, it helps to better predict commissions, and if specific actions are taken based on the tool’s suggestions, it can also boost agent confidence in re-approaching contacts that may have cooled off.
Pretty much every real estate deal, be it lease or purchase, ends with several legal documents — realty/listing contracts, property descriptions, offer/acceptance, purchase and sale, leases, and so forth.
Creating and routing those docs is standard, repeatable, and often time-consuming. While this may not require AI, there are certainly opportunities to make document workflow management more intelligent. One might use AI to evaluate documents for unacceptable legal risks, verify the legality of specific contract terms, or highlight hidden negotiation opportunities before signing.
Real estate CRM systems that use advanced AI capabilities such as robotic process automation (RPA) can also automate tasks like generating property reports, managing land records, extracting certificates, preparing litigation papers, and more. AI in real estate CRMs can also validate documents for signatures, verifications, seals, and correctness, making deal closures speedier and easier for agents.
People who have had a good experience in closing a property deal with you are likely to return—but it may not be as soon as you might wish. Subtly staying on their radar is a great way to prompt referral business to drive more revenue.
Automated “dripping” of personalized messaging and targeted content to prospects by email or text message can occur without AI. But incorporating it in a real estate CRM can make this process more intelligent and dynamic. It’s relatively simpler to prompt an agent to send a happy birthday or anniversary message to a client, but how much more valuable would it be to send them an alert when they should consider refinancing a deal because interest rates have fallen below a certain threshold?
Agents who focus on building and maintaining goodwill relationships with clients over time, improve their chances of earning repeat and referral business, ultimately driving more revenue to the agency and more commissions to themselves.
Realty agencies that manage rental properties will see more AI coming to CRM and related proptech for property management as well. For example, property management CRMs will use various AI technologies to:
Efficiently manage the massive volumes of communications and documents flowing among tenants, owners, vendors, delivery services, repair workers, etc.
Optimize utility costs across multiple properties with varying physical profiles.
Conduct predictive maintenance to extend effective lifetimes, optimize maintenance budgets, and determine replacement timeframes for rental-property water heaters, HVAC systems, appliances, etc.
Property managers stand to benefit by increasing their capacity to scale and to manage more properties, winning market share as word travels about the frictionless experiences they provide.
Opportunities to leverage AI in real estate are legion. And the more data you feed to an AI algorithm, the smarter and more accurate it becomes over time — at least in theory. To achieve that in reality, you need two things:
A sound sales model
A steady stream of high-quality data from which the model can learn
You might have a brilliant AI model, but it can still fail if you feed it data that is incomplete, inaccurate, out-of-date, too narrow in scope, or lacking in sufficient granularity. Data “observability” is a concept you will be hearing much more about as AI continues to infiltrate sales processes in real estate CRMs. In short, it means building systems infrastructure and workforce cultures that highly value and work very hard to achieve and sustain data quality over time.
We’ll save an in-depth discussion on data observability for a different day. For now, suffice to say that real estate sales leaders wishing to exploit more AI will need to be doing some deep thinking about how they intend to promote and incentivize real estate CRM data quality for the future.
Sorry, our deep-dive didn’t help. Please try a different search term.