Screen Shot 2020-03-02 at 6.01.45 PM.png

 Managing Guest Expectations with Airbnb

Team: Lucy, Emily(me), Marvin, Lauren

Team: Lucy, Emily(me), Marvin, Lauren

As a member of Northwestern’s Bay Area Immersion Program, I worked alongside Airbnb and my team of four (pictured to the left) to improve guest expectations. Through the design thinking process, I learned how to rapidly prototype, appreciate criticism and investigate compelling research questions. My team was able to produce a new review system to build guest confidence in the quality of Airbnb’s listings.

The Problem ~ Many Visitors Escape the Site Before Booking

Screen Shot 2019-10-24 at 1.02.11 PM.png

The main problem we focused on is Airbnb’s booking conversion metric. The visual above shows search funnel stages for new visitors and previous guests.

We found that the numbers inside the dashed box are very telling of guest behaviors.

  • 42% of return visitors will look at the product detail page (PDP) with dates

  • 15% of new visitors view the PDP with dates

To the average person dates may not seem too important, but adding dates into the search criteria solidifies the action of actually booking.

  • 24% of return visitors will attempt booking

  • 14% of return visitors will ultimately book

The booking numbers for new Airbnb visitors are even lower.

  • 4% of new visitors will attempt booking

  • 1% of new visitors will ultimately book

Our Goal ~ Bring up Bookings

Screen Shot 2019-10-24 at 1.23.06 PM.png

The primary intent was to increase the booking conversion metric by 20% for return visitors, i.e. increase bookings by 20%.

The purple line shows an increase of 20% for return visitors, which would result in a 25% booking rate.

User Research and Defining the Problem Area

User Interviews

User Interviews

Journey Map of the User Experience — Identifying Painpoints

Journey Map of the User Experience — Identifying Painpoints

Through numerous user interviews and exploring the pain points in a guest’s journey, we found that guests look at reviews for credibility and to identify any red flags that are not obvious through photos or host descriptions. However, most reviews are not accurate for three main reasons:

  • Guests don’t want to harm host reputation

  • The most honest feedback is mentioned only in the private feedback section

  • Most guests don’t leave reviews and when they do, comments are very generic

Designing & Prototyping a New Review System

Recognizing the review system as one of the main pain points for users, we decided to build a new review system that included more options to improve the clarity and organization of the PDP.

Implementing 3 New Features to Reviews

Screen Shot 2020-02-05 at 9.15.40 PM.png
  • Comments and reacts — comment and thumbs up or down on reviews

  • Endorsements — endorse another guest’s review

  • Filtering by endorsements and reacts — filtering by most endorsed / most liked or disliked reviews

Screen Shot 2019-10-24 at 1.58.42 PM.png

Interactive Prototype

We created the new review system on Adobe Xd to allow users to physically interact with the prototype. We wanted to make it feel like they were leaving a review themselves.

After creating the prototype, we conducted more user testing to see what parts of the prototype worked well and what did not.

Focus group testing at Northwestern’s SF Campus

Focus group testing at Northwestern’s SF Campus

The endorsement option gives guests the ability to relate to reviews without feeling like they are leaving more repetitive feedback.
— Erin Bruns
Make it clearer that only previous guests can thumbs up and down reviews.
— Liam Lecka

Final Prototype of New Review System

A visual demonstration of a potential updated review system for Airbnb featuring new and essential components derived from user testing.

The future of Airbnb’s under-utilized review system.