Optimizing Airbnb Revenue in Dublin

Project Title: Optimizing Airbnb Revenue in Dublin

Inspiration

We were motivated to analyze the Airbnb data from Dublin due to its significant role as a global hospitality platform. Airbnb has surged in recent years due to the abundance of demand and lower prices, but in cities with stricter regulations and more spontaneous travels (tourist destinations like Dublin), the planning is much more difficult. Dublin is a great tourist city example of enhancing short-term rental profitability under Rent Pressure Zone regulations, and determining whether strategies like multi-listings are recommended due to their streamlined process and service quality, and how nationalities reveal tourist preferences in duration of stay.

What It Does

Our project analyzes the provided datasets (contacts.tsv, searches.tsv) to propose actionable policies aimed at enhancing both the quality and profitability of the Airbnb market in Dublin. By identifying gaps between guest demand and host supply, namely the lack of occupancy due to scheduling and policy reasons, we suggest improvements that could be modeled in similarly concentrated tourist cities.

How We Built It

Our analysis was conducted using Python on Deepnote. We used a variety of tools to aid our investigation and implement our graphs:

Scikit-learn (sklearn) for building machine learning models to predict booking behaviors and host acceptance patterns.
Matplotlib and Altair for creating compelling visualizations to depict demand trends, booking rates, and other relevant metrics.
Seaborn for more in-depth statistical visualizations that helped us understand complex data relationships and distributions.

Challenges We Ran Into

Initially, we faced difficulties in defining a clear set of problems and solutions for our project because there were many potential paths to optimize, which led to some confusion in the early stages of data analysis. Near the completion of the project, we struggled with condensing our findings and recommendations into a presentation format that fit within the given time constraints.

Accomplishments That We're Proud Of

We are particularly proud of the specificity of policies we developed to tackle the challenges faced by Airbnb hosts and guests in Dublin. By identifying specific problems, justifying with data, and evaluate existing policies under Dublin's context, we have the potential for real-world application, provided they are supported by further market analysis and stakeholder buy-in.

What We Learned

This project was an invaluable learning experience for our team, particularly in mastering new data analysis tools and techniques. Members who were less familiar with Python libraries such as sklearn and seaborn gained hands-on experience. We also learned how to effectively collaborate on a data-driven project, distributing tasks and integrating our findings into a cohesive analysis.

Enhanced Statistical Insights from Data Visualizations

Seasonal and Weekly Booking Trends

Bookings peak on Fridays and Saturdays, with March (spring break) seeing an increase in bookings. The data is skewed towards October's months, yet many plan ahead — especially those traveling from far, like US, CA, and AU. Most tourists and local, and events like the Dublin Beatles Festival showed a 45% spike in bookings.

Impact of Rent Pressure Zones on Bookings

Listings in Rent Pressure Zones show 20% lower occupancy. A slight increase in RPZ cap could boost annual revenues by up to 10%.

Multiple Listings Analysis

Hosts with multiple listings command half of the bookings. This has been a technique in recent ten years of Airbnb markets. We analyzed our dataset to find out whether this was an effective strategy — spoilers: it is. We have included the relative probability distribution (modeled after normalizing the data, distribution of number of houses in multi-listers and single listers, plotted against the random variable of days booked).

Tourist Demographics and Preferences

Cluster Analysis: 60% of weekend bookings come from European tourists. Long-haul tourists have longer stays but lower frequency, supporting tailored experiences.

Effectiveness of Proposed Solutions

Simulations and Predictive Models: Proposed Weekend Optimizer Bonus Program could increase weekend bookings. Spontaneity: To combat spontaneity of travels and vacancies, we’ve allocated a portion of marketing budget to fund the Weekend Optimizer Bonus Program. Tiered Rewards System: Implement a tiered system where the bonus increases with each consecutive weekend booking accepted. For example, a 5% bonus for the first weekend booking accepted, increasing to 10% for the second consecutive weekend, and capping at 15% for three or more. Flexibility Reward: additional incentives for hosts who demonstrate flexibility with last-minute bookings or changes in booking dates that lead to filling weekend vacancies Weekend Ratings: On airbnb platforms, show hosts who have maintained a high weekend ratings. This is fitting for Dublin tourists, many of whom are local and looking at weekend trips.

What's Next for Optimizing Airbnb Revenue in Dublin

We plan to further analyze financial impacts of modifying rent pressure zone regulations and evaluate long-term effects of our proposed changes. Ongoing refinement of our data models and policies based on feedback will aim to create a more sustainable and profitable Airbnb ecosystem in Dublin.