Skip to main content

Airbnb is using machine learning to help hosts make as much money as possible

Airbnb is using machine learning to help hosts make as much money as possible

Share this story

If you go on vacation, regardless of whether you're staying at a hotel or renting out someone's house using a service like Airbnb, you're probably aware of the fact that what you pay is going to be based on demand. Trying to find somewhere to stay in Austin during SXSW is naturally going to cost more than it does the rest of the year. But today we learned just how good Airbnb is at helping hosts get as much money as they can — and it's all based on the back of some pretty massive data stores.

Today at the Airbnb's Open Air event in San Francisco, VP of engineering Mike Curtis (pictured above) introduced Price Tips, a tool that lets Airbnb hosts see exactly where they should set the price of their property on a day-by-day basis to make it most likely to be rented. When Price Tips fully rolls out (Curtis says it is live as of today, but a press release called it a gradual rollout), hosts will be able to see a calendar that shows the price they have set for their property every day.

Big data helping Airbnb hosts make more money

If you have it priced just right, those dates will show up green on the calendar — but if the price is too high, Airbnb Price Tips will show the price as red. From there, you can use a slider to adjust the price and get it right in the sweet spot — low enough that it'll be likely to be rented, but high enough that you're not leaving money on the table. It's like surge pricing for Airbnb rentals.

Airbnb price tips

Price Tips are based on the massive amount of data Airbnb receives and processes using the company's new, open source "Aerosolve" machine learning tools. There's a host of different factors that go into the price tip model, including listing type, location, price, availability, and how far away each date is from the current time. "There are hundreds of signals that go into the model for each recommendation," says Curtis. Airbnb says that this data is updated every single day — so if you're a busy host, you'll probably want to check out your calendar every single day to make sure you're adjusting your prices accordingly.

Of course, Airbnb has a big incentive to share this data with its hosts and help them maximize the revenue they make when they rent their property. The company's business model is based on taking a cut of rental revenue — so if it can help hosts rent their places out more often or at a higher price point, it'll benefit everyone. Except for Airbnb users hoping to find a surprise good deal. In testing, it's been effective at its goal thus far — Curtis said that a property is four times more likely to be booked if a host sets their price within 5 percent of the recommended price tip.

Airbnb graph