How Ford democratizes the use of data throughout the enterprise

The Ford Global Data Insights and Analytics team expands data access and analysis capabilities throughout the enterprise.
Written by Doug Henschen, Contributing Writer
ford logo

With its data growing exponentially and digital transformations including mobility, continuous connectivity, and autonomous vehicles quickly emerging, Ford recognized back in 2014 that it needs to take a more comprehensive and strategic approach to data-driven decision making. These were key reasons why Ford hired its first global chief data and analytics officer, Paul Ballew, and formed the Ford Global Data Insights and Analytics (GDIA) unit in January 2015.

A veteran of Dun & Bradstreet, Nationwide Insurance, General Motors, and J.D. Power and Associates, Ballew's challenge was to "take big data and analytics to the next level inside Ford... establishing an enterprise-wide vision for analytics and integrating all research, analytics, processes, standards, tools and partner engagement," stated a company press release.

Analytics tools, methods, and processes were in use throughout the company, but "it wasn't efficient to have individual pockets of the business going about analytics in inconsistent ways," Adam Blacke, lead data scientist, recently told me.

GDIA was created to share best practices and drive optimized, data-driven decision-making across the organization. Drawing on a mix of Ford veterans formerly embedded within departments as well as new hires, GDIA has grown to a staff of more than 600. Through consultative engagements, it has supported all aspects of the business, from manufacturing, research and development, and supply chain to marketing, customer service, administrative, legal, and accounting teams.

Ford analytics case study

As I explain in the case study, Ford Analytics Team Democratizes Data-Driven Analysis, GDIA's centralized coordination promotes consistency and sharing of best practices. "We knew we could learn from each other," said Blacke. "Previously, we had individual teams learning different things, but they weren't sharing across pockets of analytic exploration."

Centralization is one way organizations are making the most of available data science expertise, but there's more than one way to achieve centralized oversight. Facebook, for example, developed a hybrid approach whereby data science experts remained embedded within specific business areas, but as I explained in this 2013 article, they also reported to then-Chief Analytics Officer Ken Rudin (who has since joined Google), and they met regularly with their peers from other business units.

In the hybrid approach, experts develop deep expertise in one business domain and are always available to (and are funded by) that group. They also regularly share what they are working on with their analytics peers and trade ideas and lessons learned across business units. The chief analytics officer promotes the development of talent, sets and coordinates analytic priorities, and champions infrastructure and data investments to the benefit of all business units.

Ford's approach to centralization is equally valid and it does not prevent individual business units from retaining dedicated analytical resources. The centralized team approach is particularly beneficial in spreading data-driven decision making and optimization to departments and business units that are too small or otherwise ill- equipped to support analytics initiatives on their own.

Analytics expert Thomas Dinsmore, author ofDisruptive Analytics (one of the best books I've read on the topic of analytics), recently told me that data scientists should be handled like commercial airliners: They should always be overbooked. Centralization is one way to ensure that there's always a steady, prioritized pipeline of analytics projects to tackle. But you can also get into trouble with centralization if the queue for analytics support gets too long or if there's a lack of entrepreneurial innovation. Dinsmore said he's seen organizations go back to more decentralized (or, perhaps, hybrid) approaches after giving centralization a try.

For more information about how Ford GDIA is organized and how it promotes democratized data access with tools from vendors including Alteryx, Qlik, and Tableau, read the previously mentioned case study. Alan Jacobson, director of global analytics, explains how Ford helps users ask the right questions and choose the right tools. The report also details how GDIA tackled three projects where data science expertise made the difference in logistics, purchasing, manufacturing, and supply chain management. Click here to download a free excerpt of the 14-page report.

Other stories:

Editorial standards