Cloud Cruiser

The app that sees the customer, not just the reservation.

Comment

Cloud Cruiser

Inspiration

We started with a simple but profound question: Why are human sales agents at SIXT still better than the app?

After analyzing sources like Reddit, Trustpilot, Check24, and interviewing actual agents, the answer became clear: Context. A human agent looks at a customer and immediately knows, "This is a family with a dog; they need the Van and peace of mind protection."

The app, however, is blind. It offers everyone the same "BMW 3 Series or similar." We wanted to change that. We wanted to build an app that sees the customer, not just the reservation.

What it does

Cloud Cruiser is the first Digital Agent that doesn't just check your reservation—it checks you.

While the customer sleeps, our backend performs a "Digital Background Check" by analyzing public data (LinkedIn for profession/purchasing power, Instagram for lifestyle). It uses this data to tag the user (e.g., "F1 Fan," "High Earner," "Clumsy") and completely transforms the booking UI to match their personality.

Instead of a generic list, it offers:

The Hero Car: A specific recommendation (e.g., a Ford Mustang) based on hobbies like racing.

Hyper-Personalized Copy: Text that speaks to the user's specific interests (e.g., "Treat the F1 Fan in you").

Targeted Upsells: Insurance sold as "Anxiety Relief" based on past behavior (e.g., knowing a user recently broke their phone).

How we built it

Data Scraper & Tagger: We built a backend that scrapes public data sources to generate a JSON output containing critical tags:

Identity: e.g., Works in Tech (High Earner).

Lifestyle: e.g., Photos of hiking and lakes.

Risk Profile: e.g., Posts about breaking items (tagged as "High Risk / Clumsy").

AI Content Generation:

Dynamic Backgrounds: The app uses AI to generate atmospheric backgrounds that match the destination and season (e.g., "Munich in November—moody and cold") because context sells.

Dynamic Copywriting: The text changes based on the tags. For a hiking fan, it highlights "Winter Tires" and the engine specs for mountain driving.

Challenges we ran into

The "Creepiness" Factor: Balancing hyper-personalization with user comfort was a major topic of discussion. As we jokingly noted, "Is it a little creepy? Maybe. Is it legal? It works in the US, so it’s fine."

Contextual Interpretation: Moving from raw data (an Instagram post) to a meaningful sales tag (interpreting a broken phone post as a need for comprehensive insurance) required fine-tuning our logic.

Accomplishments that we're proud of

Atmospheric Immersion: Successfully generating AI backgrounds that match the real-time weather and mood of the destination rather than using generic stock photos.

What we learned

We learned that context is king. The reason human agents sell better is that they adapt their pitch to the person standing in front of them. By equipping our digital agent with that same level of insight, we can turn a basic rental into a personality match.

What's next for Cloud Cruiser

We plan to expand the "Digital Background Check" to include even more data points to make the experience seamless and the upsells irresistible.

Built With

Share this project:

Updates