OUTFRONT accelerates revenue forecasting and marketing campaign reporting

“Fivetran and dbt give us a consistent way to move, transform, and govern operational data so our business teams can access trusted insights faster. That foundation improves forecasting, campaign delivery, and profitability reporting while preparing us for future AI initiatives."
— Manish Gupta, VP of Data and Analytics, OUTFRONT

OUTFRONT Media is one of the largest and most trusted IRL media companies in North America, operating billboard and transit advertising across approximately 120 U.S. markets, including every one of the nation's 25 largest media markets. As part of a company-wide transformation, OUTFRONT set out to modernize its technology stack and reduce reliance on legacy systems.

The company's core operational platform was a homegrown order management system built on IBM DB2 for i more than 25 years ago, which had become increasingly difficult to scale.

Analytics relied on nightly batch extractions into Oracle Analytics Cloud and Oracle Autonomous Data Warehouse. Adding new data sources required coordination across multiple teams, schema changes were largely manual, and delivering new data sets often took weeks. For sales teams, that often meant waiting until the next day to understand pacing and revenue performance. Across the business, finance and ad operations teams also had to pull proof-of-play, contract, and profitability data from multiple systems before they could analyze performance. As demand for analytics grew, the existing architecture became a bottleneck for reporting and modernization efforts.

At the same time, OUTFRONT was working to reduce dependence on Oracle and create a more flexible data foundation that could support new analytics and AI use cases.

Building an Open Data Infrastructure with Fivetran and dbt

OUTFRONT adopted Fivetran, dbt, Amazon S3, and Snowflake to support this transformation.

Fivetran became the foundation for data movement, replicating operational data from DB2, Salesforce, and ServiceNow into Apache Iceberg™ tables stored in Amazon S3. By combining automated change data capture (CDC) with direct replication into S3, OUTFRONT moved from nightly batch updates to 15-minute refreshes across all 3 sources.

Using Fivetran Managed Data Lake Service, OUTFRONT established a single source of truth in S3. Data is stored as Apache Iceberg tables and can be accessed by multiple compute engines, including Snowflake and Amazon Athena. This architecture helps OUTFRONT avoid platform lock-in and gives the company flexibility to evolve its analytics and AI stack as business needs change.

From there, OUTFRONT queries the Iceberg tables directly or loads them into Snowflake as needed, where dbt transforms the data through a medallion architecture, creating governed data sets and standardized definitions for revenue, inventory, campaign performance, and operational reporting.

By standardizing business logic in dbt, sales, finance, and ad operations teams work from consistent definitions of key business metrics, improving trust in reporting across the organization.

Faster decisions, greater agility, and a foundation for future growth

With near real-time data now available, OUTFRONT’s sales, finance, and ad operations teams have faster access to the data they use to manage forecasting, campaign delivery, proof-of-play reporting, and business performance.

The modernization also dramatically accelerated analytics delivery. New data sources that previously required weeks of coordination can now be onboarded in hours, allowing the data team to respond more quickly to business needs. In one recent example, the team added 10–15 new DB2 tables, synchronized 24 million rows, and began building downstream data models in a single day.

Business teams across the company are already seeing the impact of more accessible, trusted data:

• Sales teams gained near real-time visibility into pacing and revenue performance, enabling faster forecasting, inventory planning, and revenue management.
• Ad operations teams centralized proof-of-play, contract, and campaign delivery data, making it easier to verify campaign execution and resolve delivery issues.
• Finance teams gained faster access to revenue and profitability insights, accelerating financial reporting and operational analysis.
• Leadership teams gained a scalable reporting foundation as OUTFRONT accelerated the migration of nearly 200 dashboards to its modern analytics platform, improving visibility into business performance as the company modernizes beyond Oracle.
• Data engineers shifted their focus from onboarding data sources and maintaining pipelines to higher-value work such as data modeling, governance, dashboard modernization, and analytics enablement. This allowed a lean team of roughly 14 people to support enterprise-wide modernization efforts more efficiently.

Looking ahead, OUTFRONT plans to build on this foundation with AI-assisted data engineering using Amazon Q to accelerate dbt development, Bedrock-powered agentic workflows to automate repetitive migration tasks, and AI-driven translation of legacy SQL, DB2, and RPG code. The company is also evaluating ways to bring programmatic advertising, proof-of-play, and partner platform data into its architecture to support more advanced analytics and automation.

Apache Iceberg is a trademark of the Apache Software Foundation.

[CTA_MODULE]