How Opendoor transformed business intelligence with Amazon QuickSight

This is a guest post coauthored with Felix Yeung from Opendoor.

Enabling self-service access to real-time, accurate insights is essential for any data-driven company. This includes selecting the appropriate business intelligence (BI) tool, which forms a foundational part of your analytics and data strategy. The right BI tool reinforces desired user behaviors, design patterns, and overall data culture.

Opendoor is revolutionizing real estate transactions through its trusted platform. Traditional home sales typically involve a 90-day ordeal of listings, open houses, repairs, and negotiations. Opendoor simplifies this into a quick, seamless process where sellers receive an estimated cash offer by simply entering their address online and providing a few details about their home. As part of the Analytics & Data Platforms team, we build and maintain the data infrastructure, tools, and products that power these core experiences and business activities.

In this post, Opendoor shares how they transformed their business intelligence capabilities by migrating from a legacy SQL-centric tool to Amazon QuickSight, resulting in 80% cost savings, faster dashboard performance, and increased data accessibility across their organization while enabling self-service analytics for non-technical users.

The challenge: Outgrowing our legacy BI tool

At its core, Opendoor relies heavily on data to make decisions and operate across markets. However, our existing BI infrastructure was creating barriers to our analysis. The SQL-centric tool catered to only about 100 technical users, effectively excluding 90% of the organization from accessing vital data across 20 departments. This limitation was particularly frustrating for operational analysts who needed to frequently analyze and slice data for decision-making. Performance issues further compounded these challenges. Simple dashboard modifications, such as applying basic filters, would trigger full query executions that took minutes to load. This computational model not only hindered real-time decision-making, but also proved financially burdensome, because the row-based billing structure resulted in tens of thousands of dollars in overage charges every few months. Additionally, the system’s fragmented semantic layer meant each dashboard operated from its own SQL query, creating inconsistencies in data interpretation and making it difficult to maintain a single source of truth. These challenges weren’t merely technical inconveniences—they were actively impeding Opendoor’s ability to scale and make data-driven decisions efficiently across the organization.

Solution overview

After evaluating several BI platforms, we chose QuickSight. QuickSight is a serverless, embeddable, and ML-powered BI service that makes it straightforward for you to deliver insights to everyone in your organization while aligning with our need for a user-friendly, configurable solution that promotes self-service analytics. Three key factors influenced our decision:

• User experience – The SPICE (Super-fast, Parallel, In-memory Calculation Engine) dataset feature in QuickSight played a crucial role in our decision. With data loaded in SPICE, executive leaders can analyze business-critical dashboards in seconds rather than minutes, transforming what was once a cumbersome process into a fluid experience. This performance improvement particularly benefits executives who need to quickly slice data across Opendoor’s key business dimensions. Furthermore, the intuitive drag-and-drop interface empowers team members to create and modify dashboards without extensive SQL knowledge, promoting self-serve analytics across the organization. The combination of rapid rendering and intuitive controls has significantly enhanced our ability to make data-driven decisions efficiently.
• Cost-effectiveness – The QuickSight pricing model was not only competitive, but allowed us to bundle it with our existing AWS services, providing 80% cost savings and a more integrated solution.
• AI-powered future – We were excited about Amazon Q in QuickSight features, which can enhance our analytical capabilities with natural language processing and machine learning (ML).

Given our team’s familiarity with AWS tools and functionality, QuickSight emerged as the natural choice for our BI needs.

Opendoor’s data infrastructure integrates multiple specialized tools to deliver comprehensive analytics. External and internal event data is initially ingested into PostgreSQL, then transformed through dbt into entity data models. These models are loaded into Snowflake for ad hoc analysis, and Databricks supports our ML research. To maintain organization across our extensive data assets, we use Selectstar as our data catalog. QuickSight sits at the frontend of this stack, pulling data primarily from Snowflake and Amazon Relational Database Service (Amazon RDS) for PostgreSQL to power our dashboards. The entire system is secured through Okta single sign-on (SSO), providing seamless yet controlled access to our analytics platform. The following diagram illustrates this architecture.

Implementing QuickSight: A swift transition

Our team embarked on an ambitious migration project in April 2024 to transfer more than 900 dashboards to QuickSight. Thanks to the platform’s user-friendly interface and our team’s dedication, we completed the migration in just five months. This swift transition allowed us to quickly take advantage of advanced QuickSight features and realize the benefits of our new BI solution. For more details about our migration journey, refer to our AWS re:Invent 2024 session Amazon QuickSight: Unified pixel-perfect reporting and dashboards.

Today, we use QuickSight in various ways across Opendoor:

• Executive reporting – Critical weekly metrics are now sent directly to our executives through QuickSight, making sure that leadership has access to the most up-to-date information for strategic decision-making.
• Self-service analytics – Opendoor’s centralized data science team builds 90% of the organization’s sanctioned dashboards across departments. This approach has enabled true self-service analytics, resulting in a 20% increase in participation from non-technical cross-functional teams who can now securely access key data related to their workstreams.
• External embedding – We’ve integrated QuickSight dashboards into our partnership portal, allowing brokers and agents to access relevant data about their workstreams. This has streamlined our collaboration with partners, alleviating the need for weekly business meetings or manual report generation.
• External report distribution – The sharing capabilities of Amazon QuickSight Pixel-perfect Reports have simplified our process of distributing data to external stakeholders.

The following screenshot shows an example of our dashboard tracking key metrics.

The following screenshot shows an example of our market metrics dashboard.

Realizing the benefits

Since implementing QuickSight, we’ve seen remarkable improvements across our organization:

• Significant cost reduction – Our BI-related costs have decreased by an impressive 80%, freeing up resources for other critical initiatives.
• Enhanced performance – Business-critical dashboards that once took minutes to load now render in seconds, allowing for quicker decision-making and analysis.
• Wider adoption – The user-friendly interface has enabled us to expand our analytics creator usage threefold, democratizing data access across Opendoor.
• Secure partner collaboration – We can now securely share relevant data with our partners through the embedded dashboards in our external portal, enhancing transparency and collaboration.

Looking ahead: Our QuickSight roadmap

As we continue to use QuickSight, we’re excited about several upcoming initiatives:

• Blessed datasets – We’re investing in creating curated, “blessed” datasets that will allow users to quickly find the most important information for their analyses. This will further streamline our data access and provide consistency across reports.
• Natural language querying – We’re eagerly anticipating the implementation of Amazon Q in QuickSight, which will empower our users to query our data using natural English language. This feature will make data exploration even more accessible to non-technical users.
• Expanding external embedding – We’re exploring ways to extend our embedded analytics to more partners and potentially to our customers, always keeping security and data privacy at the forefront.

Conclusion

Our journey with QuickSight has transformed how we approach BI at Opendoor. In just 5 months, we’ve gone from a system that was holding us back to one that’s propelling us forward. The combination of improved performance, reduced costs, and increased accessibility has empowered our teams to make faster, more informed decisions. As we continue to innovate in the real estate market, we believe QuickSight will play a crucial role in helping us understand our data, identify trends, and seize opportunities. We’re excited about the future possibilities, especially with the integration of AI-powered features like Amazon Q in QuickSight. For other organizations looking to modernize their BI capabilities, our experience with QuickSight demonstrates the potential for rapid, impactful transformation. By choosing a solution that prioritizes user experience, performance, and cost-effectiveness, you can unlock the full potential of your data and drive your business forward. At Opendoor, we’re not just changing how people buy and sell homes; we’re also revolutionizing how we use data to make that process smoother, faster, and more transparent for everyone involved. With QuickSight, we’re well-equipped to continue leading the charge in the digital transformation of real estate.

To learn more about how QuickSight can empower your organization visit Amazon QuickSight and try a free trial of QuickSight today.

About the authors

Felix Yeung is a Senior Product Manager at Opendoor, where he leads the Analytics and Data Platform teams responsible for powering the company’s machine learning valuation models, pricing tools, and self-serve analytics capabilities. At Opendoor, he spearheaded the migration from Mode to Amazon QuickSight, led initiatives to reduce analytics spend, and incubated internal AI tools to accelerate data-driven decision-making across the business.

Ramon Lopez is a Principal Solutions Architect for Amazon QuickSight. With many years of experience building BI solutions and a background in accounting, he loves working with customers, creating solutions, and making world-class services. When not working, he prefers to be outdoors in the ocean or up on a mountain.

Konala McGrath is a Solutions Architect at AWS. He supports digital-centered businesses across all industries and helps them build highly scalable, cost-optimized cloud solutions. He has been with AWS for over 5 years, helping customers migrate and scale their infrastructure on AWS. In his free time, he enjoys spending time with his family and friends, and watching or playing sports.