Pushing the envelope with pricing. An interview with Ian McHenry: CEO and co-founder of Beyond Pricing
This article summarizes our podcast interview, published in two parts on the feed at https://dsianalytics.com/feed/podcast. Links to the podcasts are included at the end of this page.
For the hundreds of thousands of individuals who rent out their rooms, apartments, and houses on Airbnb, it is extremely difficult to determine how to set rates that maximize revenue. Although hotels and airlines have made progress over the past few years in techniques for pricing their perishable inventory, private rentals face the additional challenges of highly unique product and no discretionary stock for price experimentation. In this interview with Ian McHenry, we discuss pricing techniques that his company has developed to deal with these challenges.
Ian McHenry is the CEO of Beyond Pricing, which he co-founded in 2013. His company, which is used by over 50,000 listings around the world (including Ian’s own vacation home in Santa Cruz, California), provides a dynamic pricing and revenue management software for vacation rental owners. More listers on Airbnb use Beyond Pricing than use Airbnb’s own pricing tool.
Prior to founding Beyond Pricing, Ian advised large airlines and hotels around the world on revenue management and pricing. Ian holds a bachelor’s degree from Princeton University.
David: Can you tell us briefly about the history of Beyond Pricing?
Ian: I previously worked as a pricing specialist for hotels and airlines. I was with Oliver Wyman, which had the world’s largest airline pricing practice. Around 2013, we saw that about 20% of rentals were in the private vacation rental space, but pricing in this area was 20 years behind where it was for the hotel industry. My co-founder and I started offering full service product management for Airbnb rentals, and we saw that we could optimize prices in a way that led to 40% increases. We soon decided that we should focus on the pricing aspect and leave the physical property management to others. That’s when we launched Beyond Pricing as a software-as-a-service, back in 2014.
We are now approaching 350 global markets with over 50k listings using our pricing tools.
David: You’ve published figures indicating that you can help customers increase revenue by 40%. Can you elaborate on this?
Ian: It actually goes up to 50-55%, and that’ assuming the property already started out with basically normal prices. The increase you can get in any pricing situation depends on how inefficient the market is to start with. Hotels and airlines have already been optimized quite a bit, so pricing optimization with them typically leads to single figure increases, but individual Airbnb property owners were leaving a lot of money on the table during peak periods and sitting empty with overpriced units during periods of lower demand.
David: Do you see the uptake from your pricing tools changing over time?
Ian: As property owners use the tool longer, they themselves get better at pricing. However, supply in the private rental market is extremely dynamic. Unlike the hotel market, where a new hotel takes time to build, private individuals can put their homes on the market essentially overnight. For example, we’ve seen the Airbnb supply in Tokyo and Osaka increase sharply over the past year. In such a market, you’ll have a lot of downward pressure on prices from increased supply.
The model has dynamics similar to those of Uber. As there are more cars on the road, the feedback loop gets stronger. As you have more cities in your model, it becomes easier to extend your model into additional cities. You can compare demand curves for new regions to those in old regions. For example, we’ve just launched close to ski regions in Japan, and in this case we use other global ski regions to help in modeling this new region.
David: Airbnb has their own pricing tool, an open sourced tool which they call Aerosolve. Are the price recommendations from Airbnb’s tool similar to those provided by Beyond Pricing?
Ian: This tool launched about a year after we did. We are looking at a much broader data set. Airbnb’s tool only takes data from the listings that are on the Airbnb site. In addition, we feed the booking information back into our models. We also have more people using Beyond Pricing than use Airbnb’s pricing tool, and that gives us the extra insight into booking results.
It’s important to note that prices tend to be lower on Airbnb than for similar properties listed on platforms such as HomeAway. Once you start cross-listing your property, you really need to be using a tool that compares supply, demand, and pricing across multiple listing platforms.
Lastly, Airbnb has a different incentive in their pricing recommendations. They would rather have all units priced lower and increase their overall fill rates. That pushes towards underpricing of units.
David: From what I’ve seen of both tools, it strikes me that your methodology is more feedback driven and less analytically driven than that of Airbnb. Would you say that is an accurate statement?
Ian: Our methodology is very much a model-based approach, but we do indeed pull in data from everywhere. The analysis is only as good as the data you put into it, and you have to be careful with how you calibrate or everything will regress to the mean and be useless. You need to create the right sets of rules and factors. We rely heavily on feedback loops as well as on supply and demand inputs. We also need to avoid using ‘black box’ algorithms, else our users won’t trust our results.
We are constantly looking to improve our forecasting of demand, in particular, which means focusing on
- How we can catch new demand signals
- How we can catch these signals further in advance
We’ve found that the best way to predict demand is to closely monitor when demand actually begins to surge, rather than by using scheduled events to predict them.
When you’re a small company, you have the advantage of getting valuable feedback directly from the customers, which you really need to see how the model is playing out and if there is a change in the market that you need to account for. Of course those of us working at Beyond Pricing also have our own vacation units that we are renting out.
David: Considering that each accommodation is unique, not only objectively, such as in its location, size and amenities but also in how it is presented in pictures, descriptions and reviews, how do you adjust for differences when pricing?
Ian: That’s one of the most common questions we get. The difficulty is that if you price based on a comparable properties, you still don’t know if the other property was priced well. The better way is to compare booking rates for properties in similar areas. You adjust the prices until all properties in the area are booking at basically the same rate.
David: Could you picture your pricing methodologies working well in other industries, for example in pricing real estate?
Ian: Some of it could be. Quite a number of people have approached me with ideas to extend Beyond Pricing into other sectors. Practically speaking, there are constraints in time and financing with a start-up, and moving into a new industry would require significant new steps in marketing, setup and model tuning. It’s not as simple as just switching the data feeds.
David: You have a strong background in pricing airline and hotel inventory. How did you need to adjust the standard methods for pricing these perishable goods to make them suitable for the home rental industry?
Ian: The traditional pricing method for hotels and airlines is to look at fill rates now vs. this time last year. You increase price if demand is higher today than it was last year.
There are two major differences with vacation rentals. The first is that you need to maximize revenue for each owner. You can’t let the first 100 bookings be underpriced because the last few high-priced units won’t make up for it to the owners. The second is that each Airbnb property is unique, whereas for hotels and airlines, you are essentially selling multiple copies of the same room or seat.
We’ve actually leap-frogged the airline and hotel industries, who are now also starting to price individual room and seat types using independent supply and demand curves rather than the traditional method of adding surcharges for what is really a fundamentally different product.
More details of Beyond Pricing can be found on their website: https://beyondpricing.com