Home How Big Data Could Limit Super Bowl Sticker Shock

How Big Data Could Limit Super Bowl Sticker Shock

Guest author Alex Salkever is the head of product marketing and business development at Silk.co.

Andrew Kitchell is from Seattle and is the co-founder of PriceMethod, a startup that helps AirBnB and HomeAway hosts price their properties. His co-founder Joe Fraiman is from Boston. They both follow football and pondered going to the Super Bowl, but were floored by the high prices for accommodations—even though their business is all about supply and demand, which gives them a certain insight into the impact of 100,000 people abruptly descending on a city in search of an affordable place to stay.

Credit: PriceMethod

So Kitchell and Fraiman flipped their methodology around and built a simple tool to help Super Bowl attendees find cheaper last minute lodging. They took the same Big Data harvesting and categorization infrastructure they had built and, on a dime, put a new UI on the results to make it easier for the public to search for cheap accommodations—the exact opposite of their normal business helping peer-to-peer property owners charge what the market will bear.

I caught up with Kitchell to talk with him about their Super Bowl findings and how PriceMethod crawls data and builds data models that can give property owners the same pricing tools as big hotel chains. Here’s a lightly edited version of our conversation.

Leveling The Playing Field

ReadWrite: So where did the idea come from?

Andrew Kitchell: We are a data science-focused team of Y Combinator alums, and usually we help Airbnb and HomeAway listings with data-driven pricing. However, my co-founder is from Boston, and I’m from Seattle, so we thought this would be a fun time to use our data to help our fellow football fans.

RW: Tell us a little bit about how PriceMethod works.

AK: We’re trying to level the playing field for P2P (peer-to-peer) accommodations versus traditional big hotels. To do that, we need to have a good picture of the entire market including hotels and other accommodation sources.

As a base we collect data from Airbnb and HomeAway, the two biggest P2P accommodation networks. We do that several times per day. Additionally, we collect hotel price and occupancy data from multiple sources across the Internet. Primarily, we use hotel data to build a predictive pricing model for local demand. We assume that hotels, because they have very strong predictive pricing tools, are already baking in good assumptions for local demand based on their own algorithms and historical data.

We also use vacation rental and P2P property data to build a reactive pricing model. This adjusts prices based on how local demand translates into actual bookings within a neighborhood, inventory type. You need that in the P2P market because it is still somewhat unpredictable.

RW: How do you account for things like the price of inventory taken off the market?

AK: For scraped hotel and vacation rental or P2P listings, we infer the “booked price” for any day from the last observed price. We collect data from channels throughout the day, so we will observe and record any booking within, at most, 24 hours. With a linked account, we can get perfect access to booking data. However, as a first step, we can use the last observed price to inform a robust model.

How To Build A Pricing Model

RW: Your team has some deep experience in building pricing models for big financial firms in commodities and other trading markets. How do you build your pricing models for the P2P accommodations markets?

AK: Our current pricing model consists of four components. First, we base price recommendations on the average market value of similar listings. Then we make a local adjustment due to the popularity of any given neighborhood. This adjusts and improves our base pricing model.

We then apply a time-sensitive model model informed by the booking curve of the local market, taking into account time periods expected for local bookings. Lastly, we look at demand driven changes depending on the local availability of vacation rentals and hotels.

Q: So how is the Super Bowl different in terms of pricing?

A: By our calculation, at least 75% of the P2P and vacation rental market is underpriced for the Super Bowl. We’re seeing some amazing price increases for informed owners, and our favorite example of how the rest of the accommodations market is moving is captured by the fact that someone is selling a basic room for 20x their normal rate.

For the Super Bowl, we wanted to determine how hosts could price their home during a period of exceptional demand. So we actually skewed our model to analyze how much experienced P2P hosts—those with more reviews and more future bookings—were increasing prices, and how booked out these listings were at their raised prices. In some cases, owners are increasing their prices up to 15 times their normal rates, so we were able to observe bookings at this homes to discern the efficacy of these increases.

For hosts during the Super Bowl, we used this analysis to recommend a reasonable range of price increases for other homes. For travelers attending the Super Bowl, we used this same process to determine which homes were priced best in comparison to their potential value.

Let’s Talk Nerdy

RW: What does your tech stack look like? 

AK: It’s a Rails stack with a PostGres database and Reddis for caching. The whole thing is sitting on top of Amazon Web Services so we can spin up as many nodes as we need to do our crawls. We use Mechanize for a lot of our crawling and are using a combination of APIs, mobile APIs and standard Web data to fuel our system. AWS makes it very easy to get up and running. It’s almost a no brainer. It has so many tools and for the cost and the power, it’s quite amazing.

RW: For vacation rental owners that use you, how much more money can they expect to make?

AK: Our initial numbers show we are increasing their revenue by 20% to 40%. Those numbers will get better as we have a large set of customers. We can’t disclose numbers right now but this is a huge, multi-billion dollar market that is poorly addressed right now. AirBnB is adding thousands of listings per day. We’re bootstrapping right now and are going to raise money in a few months. But we’re confident the market is there. 

Lead graphic courtesy of Shutterstock

About ReadWrite’s Editorial Process

The ReadWrite Editorial policy involves closely monitoring the tech industry for major developments, new product launches, AI breakthroughs, video game releases and other newsworthy events. Editors assign relevant stories to staff writers or freelance contributors with expertise in each particular topic area. Before publication, articles go through a rigorous round of editing for accuracy, clarity, and to ensure adherence to ReadWrite's style guidelines.

Get the biggest tech headlines of the day delivered to your inbox

    By signing up, you agree to our Terms and Privacy Policy. Unsubscribe anytime.

    Tech News

    Explore the latest in tech with our Tech News. We cut through the noise for concise, relevant updates, keeping you informed about the rapidly evolving tech landscape with curated content that separates signal from noise.

    In-Depth Tech Stories

    Explore tech impact in In-Depth Stories. Narrative data journalism offers comprehensive analyses, revealing stories behind data. Understand industry trends for a deeper perspective on tech's intricate relationships with society.

    Expert Reviews

    Empower decisions with Expert Reviews, merging industry expertise and insightful analysis. Delve into tech intricacies, get the best deals, and stay ahead with our trustworthy guide to navigating the ever-changing tech market.