A good price
What makes a bargain? Is it the cheapest thing? The quality might be bad. Is it the biggest discount? Maybe the price was not right in the beginning.
In my role at Trivago, I joined a team that worked on this problem.
Challenge
This initiative had various problems to solve — from defining a broader vision to small UI iterations. Our team was small with great impact and fast-paced.
The design we were working on had the biggest impact on the company's business metrics.
My contribution
Product strategy User research Product design
The team
1 × product manager
1 × product designers
4 × engineers
1 × researcher
1 × UX writer
Year
2019
The funnel
When you compare hotels and prices — you have 2 macro steps:
-
Check which hotel
-
Check which room and deal for that hotel
For the first one, the decision is made based on location, dates, and price. The second phase is more narrow — check the room from different booking websites. Like Booking.com or Expedia.com. You don't book on Trivago.
At that time we knew Trivago had issues showing all the knowledge the company had of hotels. Our main persona are Price Sensitive people. Often the other personas were ignored.
UI elements audit
I made a design audit of all elements related to deals in our Hotel Card. Over the years many cross-functional teams created A/B tests to improve the conversion rate. And metrics for their teams. This led to inconsistencies and a lack of a unified design language.
Hotel card sections
Next, I defined macro areas of the hotel results card to define what type of content to expose. The status quo before that was to fill the available white space for any feature.
An entire line on the card allowed for combining many flag elements. Or having a better description of why it's a great deal. Obvious — always wins.
Cheapest price
A significant percentage of our users want the cheapest deal possible. The green CTA on the card is a recommended deal based on Trivago's proprietary algorithm — good value for money.
The team came up with a hypothesis to expose both types of deals. Cheapest and recommended. Then we prepared a prototype and a user research project.
Design sprints
While polishing the hotel card — we worked on improving the experience. Once a visitor decided on a hotel. How do you compare deals from many booking websites and rooms?
We organised design sprints as the decisions affected the broader Core Product department in the company. We invited "experts" that shared their perspectives from tech, product, content, and business areas.
People book rooms
Our main concept was based on hotel rooms. This fits our user's mental model. We grouped deals based on rooms. And expose detailed descriptions and images of those rooms.
User Interview
We had 10 participants for our usability test. In-house. I created a prototype in Webflow as it allowed higher fidelity of micro-interactions.
The goal was to understand if users have a clear understanding of why we recommend a certain deal. And if they can find a cheaper alternative if they want.
Positive feedback for the concept
-
Cheapest deal highlight
-
Filtering options
-
Included vs not amenities in rooms
Improvement opportunities
-
Room details on hover were missed
-
Sorting by price with room grouping was confusing
We iterated on the concept and created an A/B test to gather quantitative insights.
Business Metrics
As the project had many hypotheses. We validated them independently with A/B tests. The majority of the company metrics increased: Clicks and Clicks to Book. Also, I worked with the team on visual iterations to improve them even further.
Conclusion
Because of the criticality of the changes we were about to make. We had to respect the mentality of Trivago as a company — incremental and quantifiable iterations. We didn't build a new game-changing redesign. More often than not, this is the right path. What gets measured — gets managed.
Looking back, I would have taken advantage of the product's new testing capabilities. And tested more granular design variations at the same time.
This would have allowed us to be more efficient and take a direction quicker.