The problem with bots

Everyone is talking about the rise of chatbots in real estate and they’re all overlooking a key issue: consumer value.

Bots are being heralded as an important trend for 2017. As Chris Rediger writes in Realtor Magazine, bots are “a solution to engaging potential customers in meaningful conversation.” James Dearsley chimes in, saying the “power of the chatbot can be used as an entry point customer engagement tool for agencies.”

Bots are being positioned as a valuable tool for real estate agents. But bots face a problem of value asymmetry: what’s good for a real estate agent is not necessarily good for the consumer.

Defining value

Bots are positioned as a tool to help real estate agents. By automatically nurturing online leads, a bot saves an agent time and energy, and delivers them hot leads on a silver platter! The value goes to the real estate agent.

But what about the consumer?

If bots are going to be successful, we need consumers to actually use them. Why would a home buyer or seller use a bot? What’s the value for them?

Structurely has launched a real estate bot, Holmes, to grow and nurture leads. The web site offers a compelling vision of how this is good for agents, but I question the consumer value. Consider the following screenshot from their web site:


Compared to the search experience on Zillow:

(The “More” option is for updated kitchens.)

It’s the same thing. The value proposition for consumers is identical to conducting a search on a major property portal.

I get that bots can have conversations with potential consumers, but how much value does that actually add? If I’m searching online for a home, I really don’t want to have a conversation about it; I want results.

The trust factor

The second problem with bots—and a critical concept to understanding their eventual usage—is one of trust. Simply put, will consumers trust a bot over a human?

As Andy Soloman states, “Research has reached the point at which IBM Watson can replicate what a physician or GP is saying 90% of the time. As a patient, however, who are you going to trust more? A robot, or a real doctor sitting in front of you telling you what’s wrong?”

Buying a home is a massive transaction in someone’s life. Will consumers trust a bot for something this important? They may use them for a more efficient home search process, but I don’t believe people will ever put a bot—or technology, for that matter—on the same level as a human. That’s why we have engineers in front of our trains and pilots in our cockpits. When it really counts, we trust humans more than machines.

A bot will never replace the role of an agent—an actual human being—in the real estate process.

This leave bots with a narrow focus of utility, primarily around information-gathering and answering questions. But these are two areas where strong alternatives already exist, specifically property portals and real estate agents.

How bots can be successful

For bots to be successful (which I think they can be), we need to focus on the additional value they can provide to consumers. This value needs to exceed what consumers already get from alternatives.

I’ve previously written about the principle of quality: when launching a new product, you must provide more perceptible value than the status quo.


And remember: if you must explain your value, it’s not as great as you think.

In the case of bots, the value being provided to consumers is, at best, on-par with what they can currently get. In fact, I would argue that in many cases it is sub-par because of the limitations around language processing. I know exactly what I’m getting when I set search criteria on a portal; with a bot, I’m never quite sure it will understand me.

To use another real world analogy, let’s consider automated customer service hotlines. Would you rather navigate a maze of “press 1 for... ” and “tell us how we can help you…”, or would you rather be connected to an actual human being? These hotlines are good for businesses, but are they good for consumers?

I believe in the long-term promise of chatbots. But for bots to provide true utility, we need to change the conversation.

So whether you’re an agent thinking about using a bot in your business, or a tech vendor trying to sell bot technology to real estate agents, please, let’s start talking about how bots are good for consumers, not just agents.

3 problems with VR and real estate

Last week Inman posted an article titled, “Sotheby’s International Realty beats Zillow and to 3-D,” suggesting that Sotheby’s had somehow one-upped the top two property portals in the US with an exciting new technology.

Congratulations, Sotheby’s. You won a game that no one else was playing.

Everyone talks about the product, but there are three fundamental business issues with 3D and virtual reality in real estate that no one is talking about.

1. Agents don’t want it

Real estate agents don’t want to disintermediate themselves, and that’s exactly what this technology does. Agents want prospective buyers to contact them, ask questions about a property, and come to an open home. That’s how they get their leads!

Agents are a central part of the real estate transaction process. Why would they encourage buyers to use a technology that cuts them out of the process? High quality content like virtual reality tours empowers consumers by giving them more information about a home without the use of an agent.

There’s a fundamental issue with VR that goes against the grain of a traditional real estate agent’s job. If agents don’t want it, traction will be hamstrung.

2. Who pays for it?

VR tours and 3D models are expensive. As the article states, Matterport and other big providers charge between $100 and $200 to capture a typical home. While it’s a relatively modest fee for a $300,000 home, when you start multiplying that by hundreds or thousands of homes, it gets expensive. Fast.

There’s also a huge value misalignment. VR improves the buyer experience, but we’re asking sellers to pay for it.

So who pays for it? Buyers? Sellers? Agents? Brokerages? Property portals? Nope. No one wants to, because of the next point...

3. It won’t sell more homes

There is no convincing answer to the question, “Will VR sell more houses?” The current process works pretty well: people find a house, look at some photos, visit the property, then decide to make an offer.

Offering a virtual reality tour or interactive 3D model doesn’t change the process. The vast majority of buyers are still going to want to visit a house in person.

VR just adds another step to an already efficient process. It’s not a 10x improvement to the process of selling a house, and it won’t replace any existing parts of the process. It’s “interesting,” “neat,” and “cool,” but it’s not effective at selling more houses.

(As an aside, I believe there is very real opportunity for VR in the new developments and new construction space. In that scenario, the addition of VR adds a very real value in the process by allowing buyers to see something they previously couldn’t. It’s adding a real improvement to the process.)

The video problem

If we want to understand the potential impact of VR on real estate, we just have to look at video.

Video uptake on the big property portals is very low. REA Group in Australia has the highest numbers I’ve seen, with anywhere between 10 percent and 15 percent of listings having a video tour. But that’s the outlier.

If you look at the other big portals, like Zillow, Zoopa in the UK, or Trade Me in New Zealand, you’re lucky to find 1 percent of listings with embedded video tours. And Rightmove, the clear market leader in the UK, has basically none!

If that’s uptake for video (which has been around on our phones for a decade) - arguably a bigger bang for the value buck - what’s the hope for VR? A fraction of a fraction of a percent.

The one value of virtual reality

Sotheby’s didn’t beat anybody to anything. The big property portals around the world are more than capable of supporting features like VR and 3D. They just chose not to. Why? Because at the end of the day, VR doesn’t add enough value, faces structural impediments to widespread adoption, and is expensive.

At this point, VR in real estate is valuable only as a PR gimmick -- which worked, because I just wrote an article about it!