Artificial intelligence (AI) brings disruptive innovation across industries, but will it ever transform real estate deals as we know them? Opinions vary, and since real estate investing is five to ten years behind the curve on trends in digital marketing (and adoption of software in general), we will have to wait for a conclusive answer to this question.
Nevertheless, real estate businesses that have implemented different types of PropTech are already seeing how technology can transform their workflow. Potential applications of AI are simply the next step.
This is why we’ve compiled a list of the best AI tools for real estate investors that are being more embraced by the day. We’ll review various tools that help investors automate their workflow and save valuable time and energy, as well as how AI can be used in daily business operations, including lead generation and property valuation.
But first, let’s cover the basics.
You are probably using up to a dozen types of technology and software that can be labeled as PropTech (virtual reality software, drones, digital maps, data services, smart technology, etc.). They make life easier for investors and assist with:
and many other tasks within your regular business operation. As much as these apps, calculators, platforms, and data services speed up the investing process, they aren’t really AI.
When people talk about AI, they usually refer to one of three types of AI: algorithms, machine learning, and deep learning. Let’s explain.
Algorithms are practically programmed to perform a specific function, “if x then y” kind of programs. A piece of code determines the next step in the processing of data.
Example: a laundry machine.
Machine learning is when a program adapts its data processing based on input in predetermined metrics. So, in a way, the software “learns” from being exposed to new information. However, this “learning” is directed toward reaching a specific goal: “get more efficient in delivering a desired outcome” kind of program.
Example: the code that decides which post to appear in your social media feed.
Deep learning is an advanced form of machine learning. These programs are capable of analyzing data to which they were not exposed before (no such metric in the code), hence the term artificial intelligence.
Example: a game bot that improves as it plays the game (a strategic game like chess, or a more complicated one).
We are not splitting hairs on AI here for the sake of it. Labeling software as AI is trendy, so do pay attention to the apps and programs you use. Don’t let someone up-sale you a pocket calculator as if it’s AI software. If you fail to note the difference, you will not only suffer financially, but you also won’t get the results you expect.
There is nothing wrong with using a simple algorithm-based program to streamline your real estate investing process (by the way, these are great for CRM systems), but if you go for a more complex application, make sure you are getting the right program.
And you don’t even need to have a technical mind to do it. In fact, you probably use applications that employ machine learning on a daily basis. Social media algorithms are a perfect example of this. These algorithms are created to maximize two metrics, user engagement and dwell time, which is why social media platforms deliver content based on your past online behavior. If you like, share, subscribe, comment, repost (retweet), and if posts keep your attention for a longer period relative to other posts in your feed – you will get similar posts next time you use the network.
That being said, let’s run through specific uses of AI in real estate investing.
Since AI tools are making their way into regular business operations of real estate companies, let’s cover the use of AI for lead generation, lead management (CRM systems), property data, market prediction, underwriting loans, and increasing efficiency.
This is a piece of software that can help real estate investors with both lead generation and lead management. In a nutshell, predictive lead scoring (PLS) is a machine learning algorithm that allows you to single out prospects who are more likely to convert (and eventually sell a house to you) based on their online behavior as website visitors.
The underlying idea is to qualify potential leads so that you can prioritize your time and effort on hot leads. Obviously, predictive lead scoring is complex and requires developing models, tracking performance, integration with other systems you use, as well as other PLS tweaks. Also, note that we said prospects are “likely” to convert – in the end, the decision to fill in your lead capture form (or respond to other call to action) is theirs to make, and they might go elsewhere.
In a way, PLS originates from predictive index tests. These are behavioral assessment tests used in the recruitment process to match employees with the job they are most likely to excel at. The predictive index has been in use for some time, and even investors use it to build a winning team. While the focus of predictive index tests is on the behavior and aptitude of potential employees, predictive lead scoring directs the same concept to potential leads (i.e. to gauge motivation of leads).
How does predictive lead scoring work in the real world? Online behavior of website visitors can be indicative of their level of motivation. For example, it can help you distinguish between people who list their homes just to see how much they’d get for them in the current market and those who are really motivated to sell. Or between people who browse through listings of homes they can’t afford and serious house buyers. Of course, the tell-tale signs will be expressed through metrics (i.e., predictive lead scoring can’t read minds, it can only monitor behavior). We are yet to see the utility of predictive lead scoring for qualifying real estate investing leads.
A lot of motivated seller leads fall through the cracks and don’t end up as closed deals because of inadequate follow-up procedures. We’ve discussed this before, and the solution might lie in employing an algorithm in client relationship management (CRM) systems.
Algorithm-based CRM systems force acquisition team members to choose the next action that needs to be taken to close with a particular prospect. For instance, once you’ve tried to call a house seller, but didn’t manage to reach them, the system allows you to schedule a follow-up call (or other action). And it goes into specifics; the task can be assigned to a team member, you can select a channel (SMS, email, call), schedule the date and time of day for the next action, etc.
This is a very simple algorithm (compared to advanced forms of PropTech software), but it does serve to improve lead management.
Investors who use data services to compile motivated seller lists are familiar with all sorts of filters for targeting properties (and their owners). Usually, the properties are filtered based on features like number of bedrooms, square footage, location, status of the property (ex. tax delinquent houses or code violation lists). Advanced filtering, however, can include homeowner status (absentee owner, out-of-state owner), year of build, real estate market data (for that ZIP), sales records, etc. And the end goal is to either target the property owner with a marketing campaign (PPC, social, direct mail) or to use this data when placing an offer on a house.
Some of these home search tools are not really complex, but they enable real estate investors to find comparable properties within minutes, and by extension, help make a purchase decision (or at least come up with an offer on the house). Other, more sophisticated home search tools have taken the process to a whole new level. We refer to iBuyer companies like Zillow, of course. They use 1000s of algorithms programmed to perform deep learning within what is known as a neural network in AI, and then provide a property appraisal service with a median error rate below 2%. Or companies like Deal Machine which transformed traditional house hunting methods like driving for dollars with their PropTech.
The typical real estate investor could not match the AI power available to iBuyers. Nevertheless, it seems that a move toward the use of AI services for handling property data (in some shape or form) is on the horizon.
Investors try to understand real estate markets since a lot is at stake when you decide to keep a property (buy and hold strategy). It’s only natural: if you could foresee that property appreciation in a market like Boise, Idaho would enable homeowners to enjoy an increase in value of almost 400% over two decades, you’d probably want to get in on the action, wouldn’t you?
Those who have the means (i.e. iBuyers) turn to AI for answers about home prices. No one is really predicting prices, rather the network of algorithms is processing historical and recent input on sale prices, rents, property data, submarket data (a ZIP code, or type of home) to come up with an estimate. In essence, it’s data flow and projection of trends into the future with a particular metric in mind (estimate of rent growth over time in a submarket).
AI-enhanced tools for predicting property prices, like Zillow’s “Zestimate,” have a long way ahead, though. The pandemic threw a wrench in the algorithm, debilitating its capacity to project estimates more than a few months into the future, and we all witnessed the collapse of Zillow as a result.
Mortgage applications are a real nightmare for those who aren’t fond of preparing the paperwork. You simply can’t get a mortgage approved without going through heaps of documents that are required during the underwriting process.
Machine learning provides relief to real estate investors (and underwriters in general). Although input from a professional underwriter is necessary to submit a mortgage application, AI significantly speeds the process up. This is made possible by advanced text extraction and analysis tools, like optical character recognition and natural language processing. Underwriters will check the documents, but the boring and repetitive task of preparing them is handled by software.
If you pour data about your processes into an AI system, you can get an analysis as an output. Since machine learning can observe the interactions from a detached perspective (past our prejudices and learned values), it can also present novel ideas on cutting costs and increasing operational efficiency.
For example, there are AI tools that analyze the use of office space. They can help you cut costs on utilities and rent for your office by pointing out inefficient use of the space you have. Granted, not all investors have an office, but those who do have them would benefit from these applications.
This approach to analyzing efficiency can be applied to construction projects as well. For this to work, contractors need to integrate their working processes with the Internet of Things (IoT). AI tools can also help improve scheduling (noting supply chain disruptions), and all those sensors and feedback loops can spot safety issues before humans can.
Now that you know how AI can be used in real estate, let’s take a look at specific AI tools that can ease the life of investors.
We use the term “AI tools” broadly here. As you can see from the examples that follow, these AI-powered real estate investing solutions fall into different categories (digital platforms, search engines, CRM systems, data services, etc.). They include services that can be accessed by any user. Sometimes the AI part is a small component of the overall offer, other times it’s critical for the product.
This will become clearer as we check out the following AI tools.
House Canary is an AI-powered property valuation tool, but it can also forecast house prices (near term). Its analytical model is based on historical data (inputs that go 40 years back) and it boasts an error rate of below 3%, which is impressive. This tool is used for residential real estate valuation (including rental properties) and offers financing, acquisition, and data solutions to its users.
Skyline AI provides estimates on commercial real estate. It’s based in New York and Tel Aviv and provides valuation and analysis on investments in commercial property. The Skyline AI team is well versed in creating state-of-the-art AI (they’ve sold two such companies thus far), so they offer users forecasts on real estate market trends based on past transactions. The tool is used for finding great investment opportunities.
Tririga was developed by IBM as an application for ensuring efficient use of office space, or as they label it – integrated workplace management solution (IWMS). In essence, it analyzes inputs from sensors to eliminate underutilized space in office buildings. Occupants can plan use of workplace space in real-time; to book a room for a meeting, schedule relocation, and other requests.
Enodo is primarily a tool that can help you underwrite multifamily deals by using machine learning. However, it also has a robust home search tool, and you can easily find comparable properties, analyze rents (drawing on historical data), and run calculations on expenses (particularly for rents and leases). Real estate investors can use it to get an estimate of the after repair value (ARV) before they start remodeling projects, although the metrics available are not as straightforward as ARV.
Proportunity is a company that helps would-be property owners to get a fair and up-to-date valuation of residential real estate. Their model gathers property data, real estate market data, crime rates, and similar inputs to provide automated property evaluation. They also offer loans (Proportunity loans), but before you get to that step, you’d be notified whether the house you are looking to buy is overvalued or undervalued. At the moment, Proportunity is only available in the UK.
Doxel is created to track construction projects through AI. Its purpose is to help project managers stay one step ahead of delays that could extend construction deadlines, and to cut costs. Doxel employs autonomous sensors and machines to monitor progress, and its users enjoy the perk of having a visual representation of completed work (available in 3D). This AI tool serves to increase the productivity of construction teams.
Hyro provides AI assistance in conversion and lead nurture for real estate investors and property managers. Traditional chat bots deliver mixed results, however, software like Hyro adapts to requests from website visitors. It’s designed to capture leads across channels and to eliminate the need for virtual assistants or customer service teams (great for real estate investing solopreneurs). This software uses natural language understanding to improve website conversions and collects data on visitors that a representative (human) would miss. Hyro can also be used for property management and for managing relations with tenants.
Trulia helps its users to find the perfect home by offering detailed and up-to-date insights about the quality of life within submarkets (neighborhoods). So where’s the AI part? Well, AI is utilized to improve the user experience. Algorithms track even minute distinctions in online behavior, including the wall colors of the photos that kept the interest of website visitors. This input is then used to set personalized browsing criteria. Real estate investors might be looking for a way to do the same on their own sites soon.
Lofty AI is a digital platform that enables users to make micro investments. In a way, it aims to replace real estate investment trusts by offering tokens for as little as $50 to investors. These tokens are created with blockchain technology and Lofty AI practically makes it possible to crowd source your way into collecting passive income from rental properties. They calculate daily rental income to attract investors (it can be withdrawn on a daily basis) and the value of the token follows fluctuations due to property appreciation over time.
Compass is a home search tool. It’s used by thousands of real estate professionals and its AI team has hundreds of members. Compass provides a great user experience (similar to Trulia), so each visitor receives personalized results based on preferences from past online behavior. However, the distinctive feature of this tool is its AI-powered CRM system, capable of tracking the activity of individual platform users. In essence, you get invaluable info about the online behavior of potential leads (so you can act on it).
AI has the potential to change real estate investing as we know it, however, we haven’t really witnessed this transformation yet. How will this materialize in the future? It’s anybody’s guess.
There are different levels of AI, so when you get PropTech to improve your in-house processes, check whether you are buying an algorithm, machine learning software, or advanced (deep) learning software. Many types of PropTech employ AI, in one form or another. Some of the most notable apps are Skyline AI, Hyro, Trulia, Compass, House Canary, Enodo, Doxel, Tririga etc.
The application of AI in real estate investing is an emerging field and all of these apps are practically test runs. Companies like Zillow paid a high price for relying on AI too soon. While theirs is a cautionary tale, AI tools are likely to play a bigger role in this industry as time goes by.
Reputation is everything in real estate. It takes hard work to build a good reputation, and even the slightest blunder can ruin it. And, today, with social media (and the internet in general), it’s really easy to check the track record of a specific company, your real estate investing business included. To help you create
Artificial intelligence (AI) brings disruptive innovation across industries, but will it ever transform real estate deals as we know them? Opinions vary, and since real estate investing is five to ten years behind the curve on trends in digital marketing (and adoption of software in general), we will have to wait for a conclusive answer