Lew Sichelman
“REGAL HOME: VIP residence with parkside charm. Remodeled, stone-pillared colonial on 18 acres, first-owner pride, 18 fireplaces, master suite, modern kitchen, 12 BR/12 baths, pantry, finished basement, swimming pool. Near bus. Fencing, formal dining rooms, garden, wood paneling, balcony, servants’ quarters. FHA-VA $30 million.”
If there’s anything most real estate agents dread – and that most don’t do very well – it’s writing the descriptive, narrative ads for the houses they list. The description above, in case you haven’t guessed, is for the White House, and it was written by a computer in the 1980s for a company called AdWriter. It may not be particularly scintillating, but it did the trick.
Some realty pros have a knack for writing these narratives, no doubt. But for many, the task looms with all the charm of a final exam – even though telling the property’s story in an enticing manner is perhaps the most important aspect of adding the house to the local multiple listing service, where most buyers start their search.
But there is hope.
AdWriter is still around, writing copy for about a dozen companies. But “it was built for print,” said Ken Douthit, son of company founder Harold Douthit, and newspaper advertising isn’t what it used to be.
Other outfits have come around that offer a little more panache.
Restb.ai is the latest in a string of artificial intelligence-powered technologies that have launched recently, all aimed at helping agents automate time-consuming tasks like writing property narratives.
For a sample, here’s the beginning of Restb.ai’s 335-word description for the presidential mansion: “This stunning neoclassical-style building features a striking white exterior adorned with grand columns and intricate detailing. The White House comprises 132 rooms, including 35 bathrooms, spread across six levels, providing a mix of formal and informal spaces to cater to the diverse functions of the presidential office.”
Caveat Emptor
Among the new AI offerings, there are chatbots such as ChatGPT from OpenAI, Google’s Bard and Microsoft’s Bing. All work a little differently, and all leave something to be desired. At best, they may require the agent to tweak the results, and at worst, what they spit out is not worth the few seconds it takes to read.
AI is a “very powerful tool as far as being creative,” said Tony McGrory of the PTG Group. “But there are a number of downside risks.”
One is accuracy – or lack thereof. For example, ChatGPT said I wrote a book, which I didn’t. Even worse, I must be typing this from the Great Beyond because it also said, “Sadly, Lew Sichelman passed away in 2019.” It is hot here, but I assure you that the column you are reading was not written by a machine. (Thanks for the “sadly,” though.)
Back to real estate: While some descriptions may be “plausible,” McGrory told me, “they can have only an element of truth. If the information isn’t verified, you can very quickly transmit misinformation.”
The product “has to be absolutely verified,” agreed Long and Foster agent Hill Slowinski in Maryland, who uses ChatGPT and said he is satisfied with the program.
“I had to put a lot into [the narratives] to customize them,” he told me. “But it gives me a logical outline, and pointed out some things I hadn’t thought about.”
New Software Scans Photos
ChatGPT responds to questions, prompts and a list of agent-provided features. Within seconds, it culls vast amounts of data from the internet, books and articles, then produces listing descriptions.
Restb.ai, on the other hand, is not a bot, at least not in the true sense of the word, Chief Product Officer Nathan Brannen explained to me. It doesn’t chat, either.
Rather, it uses computer vision to scan photos submitted by the listing agent. It extracts details from the images, along with information from data providers like CoreLogic, Black Knight and public sources such as tax records and school districts. The technology then writes a complete description that automatically populates the agent’s particular MLS platform.
The program can identify more than 300 features, taking the description far beyond the details in a typical listing, and it can spin out write-ups in more than 50 languages. Agents can even select different styles or tones – whimsical, for example, or simple and straightforward.
If there is a drawback, it seems that the program doesn’t know when to stop. At my request, Restb.ai produced descriptions of eight famous houses – including the Hearst Castle in San Simeon, California; the Frank Lloyd Wright masterpiece Fallingwater in Mill Run, Pennsylvania; and Graceland in Memphis – and all were multiple paragraphs long.
But Brannen said the write-ups are wordy only because these properties are upscale and famous. Narratives about regular places like yours and mine wouldn’t be as verbose, he promises. Moreover, the technology can write to any word count an agent desires.
And what about mistakes?
“We’ve done a lot to reduce errors,” the Restb.ai spokesman told me. “I won’t say it’s not possible, but if there are errors, it’s likely because the data is drawn from public records.”
Of course, as with all these technologies, agents can either publish the generated stories as-is or use them as a starting point for their own text. But the whole idea is to help agents do something better while saving them time.
Lew Sichelman has been covering real estate for more than 50 years. He is a regular contributor to numerous shelter magazines and housing and housing-finance industry publications. Readers can contact him at lsichelman@aol.com.
