Automated complex manual calculations within the data model, streamlining business processes and improving efficiency
Enabled a lean analytics team to scale operations alongside organizational growth
Accelerated velocity and business impact by empowering users with self-service capabilities
Achieved all-time high organizational survey ratings for data access satisfaction
Intel Central Engineering group's Design Ecosystem Solutions Analytics team adopted Looker's central semantic model to manage complex software license usage data without creating data silos and automated complex forecasting, enabled self-service for business users, and fostered a thriving data culture.
Intel Central Engineering group's Design Ecosystem Solutions Analytics team adopted Looker's central semantic model to manage complex software license usage data without creating data silos and automated complex forecasting, enabled self-service for business users, and fostered a thriving data culture.
Innovation without abstraction
We avoided a strategy that would involve pulling data from various sources and joining it locally. That approach introduced abstraction, increased the risk of data getting out of sync, and raised technical debt concerns. We wanted to leave the data in place and centralize the model.
Divya Sripathy
Technical Program Manager, Intel
In 2017, Intel's Design Ecosystem Solutions analytics team sought a solution to maintain dashboards and forecasts for software license usage — a finite resource critical to their engineering operations. To do this, the team needed to bring together data from different groups across the company. When architecting their solution, the team explicitly avoided a "copying strategy" that would involve pulling data from these various sources into a local warehouse for reporting. They realized this approach would introduce an unwanted layer of abstraction between the original data owners and end users and increase the risk of data copies getting out of sync, reducing confidence in the data and resulting insights.
Instead, the team designed a new architecture built on three foundational tenets: leaving the data in place to maintain clear ownership and minimize abstraction; centralizing the data model to transparently define how different sources join; and leveraging a data virtualization layer to translate SQL dialects across sources on the fly.
They chose Looker for modeling because its features aligned perfectly with this vision. Looker's platform allowed the group to define a central model that was version-controlled and widely accessible for contributions, while its ability to connect to a large variety of data sources without moving the data enabled them to execute their strategy of innovation without abstraction.
Building a culture of self-service and community
Since implementing Looker, the analytics group has moved beyond simple reporting to full automation of complex tasks. A key "aha moment" occurred when the team realized they could leverage Looker's central semantic model to handle highly complex calculations that previously required manual work on a monthly or quarterly basis. By modeling these intricate equations directly in the semantic layer, they automated the process, increasing productivity by freeing up the team's bandwidth. This automation, together with other time-saving features the team built on Looker, allowed the analytics team to scale alongside the organization while staying lean. They were able to shift their focus from maintaining infrastructure to tackling new use cases and improving business processes identified by user feedback.
The central semantic model provides consistent value by accelerating the velocity and business impact for both the data team and end users. It allows our lean team to focus on building new models and tackling new use cases instead of infrastructure maintenance.
Divya Sripathy
Technical Program Manager, Intel
This technical foundation was critical for fostering a strong data culture where everyone, regardless of technical ability, could use data creatively. The central semantic layer empowered business users — the true domain experts — to explore data and build their own metrics rather than being restricted to a static set of dashboards provided by the analytics team. To ensure successful adoption, the team executed a comprehensive engagement strategy built on communication, training, and community. They developed specific training curriculums for different user types — viewers, developers, and explorers — and recruited "super users" across business groups to act as local advocates and experts. These super user forums sparked internal partnerships, allowing users from different groups to discover shared use cases and collaborate on solutions.
Additionally, the team established a structured intake process to ensure user feedback and questions were routed quickly to the right person, reinforcing a culture of support and responsiveness. This combination of robust tooling and community building has resulted in organizational survey ratings for data access satisfaction reaching an all-time high. Looking forward, the team sees this accurate and centralized data foundation as the prerequisite for leveraging AI, and they are excited about the potential of tools like Gemini to further improve the end-user experience through conversational analytics.
The Design Ecosystem Analytics Team in Intel's Central Engineering group supports internal engineering and operations organizations through centralized infrastructure, enabling reporting and analytics for design software license usage, forecasting, and spending metrics.
Industry: Technology
Location: United States
Products: Google Cloud, Looker
