The Technology Stack Behind Airbnb

When people think of billion-dollar companies, Airbnb often comes to mind. Even throughout the pandemic, when traveling has become something fantastic or even impossible, Airbnb remains in the Unicorn Club and is valued at $18 billion. And the company is still a perfect example of how the sharing economy can change the whole hotel and travel industry.

In this article, we’ll uncover some particularly interesting features of Airbnb’s underlying technologies. But first, we’ll talk about what this online house rental marketplace was like before it conquered the world.

How Airbnb got into gear

Just as Uber disrupted taxi booking services, Airbnb brought a wind of change to the hospitality market. This encouraged many entrepreneurs to create Airbnb clones. Who would’ve thought that the age-old hotel industry and the entire real estate rental industry could be disrupted by such an idea?

The strange combination of hotel and couch surfing called Air Bed & Breakfast was launched in 2008 by Brian Chesky, Joe Gebbia, and Nathan Blecharczyk. Though in the early days it seemed crazy to many, Air Bed & Breakfast, now known as Airbnb, was destined for greatness. Over 150 million Airbnb users rent spaces each year in 220 countries and more than 100,000 cities. There are over 7 million listings on the platform, and Airbnb corporate has expanded to 34 offices around the world.

Startups can often be described using Paul Graham’s “this for that” template. Airbnb used to be dubbed the “eBay for space.” Nowadays, how many startups do you know who call themselves the “Airbnb for X”? 

It’s obviously not enough to have a great idea for collaborative consumption. You also need to do something to grow your idea into a platform – you need to hack growth. In order to get their first visitors, the co-founders of Airbnb integrated with Craigslist, allowing Airbnb hosts to automatically create postings on the site. 

The Airbnb community started in New York, which was the test market for the co-founders to “do things that don’t scale.” Paul Graham’s advice became the main vision for the company. “It’s better to have 100 people love you than to have 1,000,000 people like you,” says Brian Chesky, the CEO of Airbnb.

Airbnb’s international expansion began in 2011, when they acquired the German competitor Accoleo. From that day on, Airbnb started growing like crazy and acquired Gaest.com, Luckey Homes, AdBasis, Accomable, Trooly, Deco Software, Tilt, Luxury Retreats, Trip4real, ChangeCoin, and even more companies. One of Airbnb’s latest acquisitions was HotelTonight, a company valued at $463 million! After acquiring HotelTonight, Airbnb substantially widened their offerings with rooms from the best boutique and independent hotels worldwide. 
 
How can you create an app like Airbnb? How does Airbnb work? What technology stack does Airbnb use? If you’re considering building an app like Airbnb, you’ll have to deal with major challenges that include global payments, search and discovery, messaging and communication, trust and safety, the user experience, and growth strategies. Let’s take a look at the technology stack that helped Airbnb deal with these challenges. 

Search & discovery

Airbnb uses some standard tools such as filters to help guests find the right place to stay. Users can search for accomodation by: 

  • Date

  • Number of guests

  • Listing type (private/shared room, entire house)

  • Price

  • Listing info (bedrooms, beds, bathrooms)

  • Amenities (anything from Wi-Fi to the number of fire extinguishers)

  • Instant booking or 24-hour wait for the host to accept or decline the reservation

But Airbnb wouldn’t be so popular with users if it didn’t offer the most relevant listings. How does the platform compute relevance when there are so many factors that need to be taken into account? Airbnb’s technology stack is highly scalable. To perfect their search and discovery process, Airbnb:

  • Built a sophisticated search algorithm powered by artificial intelligence to determine the relevance of locations and requirements. This algorithm combines dozens of signals to surface listings that correspond to a user’s search and ensure that listing descriptions match images provided by hosts.

  • Implemented an image processing algorithm to make the system understand whether an image is upside down or contains objects not mentioned in the listing description. For instance, if a picture shows a bath and a shower, the system will suggest that the host add the shower to the listing if it’s not noted.

  • Created an engaging discovery experience called Saved, allowing users to like search results and get back to them at any time 

To help users make informed decisions while searching for a place to stay, Airbnb has added even more goodies:

  • Experiences and restaurants to find the best places to eat or discover sophisticated ways to relax and be entertained, like, say, attending a violin concert by a local musician 

  • Airbnb Neighborhoods to let users see the places they’re planning to visit. 

  • New property categories that allow users to filter by boutique, bed and breakfast, unique space, vacation home, entire home, private room, and shared space

  • Travel themes including family, wedding, social stays, honeymoon, work, one-of-a-kind, and group party. For instance, if you choose the work theme, properties shown will supply all amenities for work and private sleeping quarters for each colleague you’re traveling with. 

  • Additional listing tiers – Airbnb Plus, Airbnb Luxe, and Beyond by Airbnb – to offer properties with verified quality and comfort or only high-grade alternatives that offer custom experiences and great hospitality 

However, demand for all of the sophisticated functionality above has shrunk during the pandemic. Aside from introducing strict standards of house sterilization and touch-free check-ins, Airbnb has had to adapt and offer something new to its users.

Online experiences became the company’s COVID-19 response, helping people find unique online activities – say, meditation with an instructor, tarot reading, or a yoga class – while staying at home. Additionally, Airbnb presented Nearby stays as another way to hold on through these tough times. This feature has turned out to be especially popular among residents of countries with closed borders. 

Airbnb web app

Read also: Eventbrite development

Messaging & communication

You might wonder what measures Airbnb takes if hosts don’t respond to requests within 24 hours. Somebody needs to ping them, right? Airbnb’s experience suggests two ways to handle this situation: call the host manually or have technology do it for you. 

Airbnb sends push notifications to hosts who haven’t responded to requests and, in case those notifications don’t work, automatically sends SMS messages using a communication API by Twilio (definitely a great tool for those who want to develop an app like Airbnb). A message to a host contains information about the potential guest, booking dates, and the price for the stay. 

To help users and hosts communicate with each other, Airbnb provides a real-time chat. It’s a comfortable tool that lets users instantly ask questions or request additional services from their hosts even after they’ve checked in. It’s also great for hosts, as providing extra services may persuade travelers to stay longer or come back again, meaning more money.

Users can text multiple hosts at once and create chats with other users while planning a trip. Hosts can not only start threads and send real-time messages but can save common responses to their current and potential clients. And both users and hosts can easily navigate these chats, which are presented as a feed with a search bar. This feed is actually more of a message center, as it’s where users and hosts see all types of messages.

Airbnb saved responses

Global payments

As Airbnb operates in 220 countries, you can only imagine how many transactions they process every day. Airbnb is licensed as a money transmitter, but it doesn’t handle payments itself. To facilitate the massive number of transactions that move through the app, they integrate with dozens of local payment providers and maintain bank accounts in several currencies.

Airbnb allows users to pay with PayPal in the US and other countries where this service is popular and has no restrictions. But for payments in other parts of the world, Airbnb uses Braintree, one of the best payment systems out there (Uber also uses Braintree). We compared Braintree with Stripe, its main competitor, and came to the conclusion that Braintree is more impressive given the wider range of payment possibilities it provides. 

Braintree is also more convenient for users, as they only have to link a card to the payment system to pay for reservations and hosts only need to do the same to withdraw money. It takes more effort to create an account and pay or withdraw money with PayPal.

And to make things even better, Airbnb has integrated with Apple Pay in their iOS app and Google Pay in their Android app. These payment options are not only easy to use but are secured by Apple and Google, respectively.

Listings on Airbnb

Transactions on Airbnb can be complex, as they may include recalculations, taxes, and fees. Moreover, each financial operation has to be associated with a definite type of payment: payment for accomodation, security deposit, coupon, etc. However, the checkout flow has to be flawless and smooth. This is why Airbnb built an interface that looks the same for all payments on the platform:  

  • First, the platform checks the product inventory.

  • Next, the user goes through a standard checkout flow.

  • Finally, the checkout is complete.

At the final stage, users see all the fees and taxes along with any additional costs. This system allows Airbnb to quickly change products on the platform while users have a consistent experience paying for them.

To help hosts set the best prices for their listings, Airbnb has implemented a dynamic pricing feature based on their Aerosolve machine learning system. This feature predicts if a listing will be booked on a certain day at a chosen price. This way hosts can optimize their pricing and get more revenue from listings.

Trust & safety

Reliable service is probably the most important criteria for users, especially if you’re developing a rental app like Airbnb. But a valid email address and phone number, which were initially the only requirements for creating an Airbnb profile, aren’t a solid foundation for trust. Now Airbnb ensures trust and safety in a number of ways:

  • Using multi-step verification to protect user accounts when entering the platform from a new device or making changes to an account

  • Using a risk scoring algorithm that analyzes traveler and host behavior to block suspicious operations on the platform 

  • Conducting background checks of hosts and travelers thanks to synchronization with official criminal and terrorist databases in the US

  • Offering masterclasses for hosts to ensure they provide high-quality services and know important information about the local area

  • Supplying smoke detectors to hosts for free

  • Automatically deleting from Airbnb’s messaging system any message that includes a user’s contact details to make sure all transactions go through the site (Users may exchange their emails and phone numbers only after a reservation is made.)

  • Allowing reviews to be written by guests only after the stay

  • Using AI to recognize the language and content of reviews in order to sort them by naturalness of the language and relevance to the listing

  • Providing Superhost certification to help travelers discover places by hosts with good reputations who have minimum 4.8 ratings, a 90% positive response rate, over ten stays per year, and no cancelations

  • Enabling Social connections by integrating with Facebook’s Graph API – which represents Facebook objects and the connections between them – to let users find places to rent from hosts who are Facebook friends or friends of friends and see if a Facebook friend has reviewed a host

  • Offering a $1,000,000 Host Guarantee, meaning Airbnb provides up to $1 million in compensation to hosts should their property be damaged by guests 

  • Placing verified image tags on professional photos uploaded by hosts

Reliability & scalability

Airbnb currently has a bit more 150 million active users worldwide, and all of them enjoy a stable and quick platform. The secret behind this reliability is a properly developed platform architecture. Over time, the company shifted from a monolithic structure to a service oriented architecture (SOA) in which a number of services are loosely coupled and can be changed without affecting each other.

Just like most high-end products, Airbnb is a cloud-based application, which allows it to quickly scale and handle heavy workloads. It’s built on the following technology stack:

  • Amazon RDS cloud database to store data and simplify administration

  • Redis for cache infrastructure and the main database

  • Amazon S3 and EBS cloud storage to store user data

  • Amazon CloudFront to instantly deliver content to users around the world

  • Amazon EC2 cloud hosting to carefully distribute incoming traffic and handle sudden traffic increases

  • Big data tools by Airpal, Presto, and Druid to store, process, analyze, and manage huge amounts of data

Monetization

Let’s have a look at Airbnb’s business model. Airbnb makes money by charging fees on bookings made through the app. Guests pay a 14.2% non-refundable fee when a reservation is booked and approved by the host.

Airbnb also collects a 3% service charge from hosts. In addition to these fees, Airbnb may charge value-added tax (VAT) depending on local and federal tax laws.

According to one study by ValuePenguin, people tend to spend more money when the payment process is easy. Following this logic, Apple Pay and Google Pay may lead users to spend more in the Airbnb app.

Airbnb doesn’t decide how much to charge for accommodation. They let hosts define the price for their properties and whether to charge any of the fees permitted on the platform, including taxes.

Read also: How to build an app like Instacart

Growth and customer experience

Airbnb positions itself as an e-hospitality brand aimed at improving the customer experience. According to Gustaf Alströmer, a former growth product manager at Airbnb, the philosophy of the company is to have users tell the story.

This is why Airbnb launched neighborhoods, included local recommendations and host guides, and even bought Localmind – a startup that knows how to get the right answers about places directly from the people that live there.

Airbnb’s growth strategy revolves around features that touch non-users: referrals, wish lists, neighborhoods, listings, references, and their amazing blog. Word of mouth, though, has proven the most efficient growth tool for the rental service. Airbnb’s daily bookings tripled when they implemented their referrals program.

Depending on the type of your stay, there’s a good chance you could save up to $100 per night staying with an Airbnb host instead of in a hotel room. The benefits of the sharing economy make the Airbnb platform a great tool for regular people to save money and for businesses to reduce their employees’ travel expenses. 

If you need any help developing a property rental mobile app like Airbnb or a fully-fledged multiplatform rental ecosystem, you know where to find a web and mobile app development company! Check out some of our works  and designs for real estate companies or view Zillow’s review of their experience working with Yalantis to see if you’ve come to the right place.

4.5/ 5.0
Article rating
242
Reviews
Remember those Facebook reactions? Well, we aren't Facebook but we love reactions too. They can give us valuable insights on how to improve what we're doing. Would you tell us how you feel about this article?
Need a technology vendor?

Discover how we can help you augment your real estate software

Contact us

We use cookies to personalize our service and to improve your experience on the website and its subdomains. We also use this information for analytics.