How an agile product approach to analytics can accelerate data-driven innovation

1


14

Agile colleagues

Despite years of investment in analytics initiatives, some organizations still struggle to deliver the right data and insights to the right people at the right time. CIOs looking to get more from their analytics efforts should consider an approach that they may already be using elsewhere in IT: agile development. Applying proven agile product management practices to analytics projects can unlock business value and accelerate data-driven innovation.

IT and business leaders striving to make their organizations more data-driven have concentrated largely on improving the level of data maturity across disciplines, such as data management or data governance, without enough focus on how data delivers value to the business. Traditional approaches to stakeholder engagement have also fallen short in fostering shared responsibility across IT and business users for measuring data’s full value.

“A lot of these early efforts were done with an IT focus and often didn’t deliver the desired business benefits,” says David Piontek, EY US-West Region Data and Analytics Leader. “Projects were directed to improving data maturity, not on how that maturity will deliver value.”

A new paradigm

Adopting an agile approach to instill a product-driven mentality across data and analytics teams can change those outcomes for the better. The concept of establishing product owners and managers encourages business users to take accountability for data products and embrace a culture of continuous development, whether for analytics models, reports or dashboards. An agile mindset also aligns well with developing ecosystems and a platform-based approach, which simplifies complexity through delivery of basic data product building blocks that people can use and reuse across the enterprise.

“The beauty of agile product management is that the measurable outcomes come in many flavors, from driving revenue and cost savings to improving adoption and reducing complexity of your data infrastructure, which inherently costs less to maintain,” Piontek explains. “It also drives more consistency in results.”

To adopt an agile approach to analytics, CIOs, chief data officers and other data analytics leaders should consider the following steps:

  • Revisit your data strategy and operating model. Given that this is an entirely new approach to analytics, it’s important to take a fresh look at existing data strategies, roadmaps and organizational structure. Data leaders need to reconsider everything from defining data products and creating data product taxonomies to re-evaluating legacy infrastructure and the makeup of existing data and analytics teams.
  • Build a data analytics product marketplace. Modeled like an app store, a marketplace for data products provides a central place where enterprise business can find data and analytics tools to address their needs. Organizing the marketplace by business function (e.g., logistics, sales, finance) makes it easier for business users to find the data products they need to drive value for their particular area or responsibilities.
  • Maintain a hyperfocus on value measurement. IT should not be the stewards for quantifying the value of a data platform and program; rather, shift to a shared responsibility model that makes business users equally accountable. This approach will enable better visibility of data-related IT investments and ultimately drive adoption of the data platform.

Adopting an agile, product-based approach to data and analytics takes time, but the payoff can be significant. “Building up this practice doesn’t come easily because it’s not a muscle most organizations have,” Piontek says. His advice: “Build the right use cases and processes up front. Figure out how you’re going to measure value. When business functions can actually see the ROI on the money they’re spending on analytics, that changes behavior.”

Read the full-length article to learn more about how companies can become data-driven organizations through agile analytics.

The views expressed by the author are not necessarily those of Ernst & Young LLP or other members of the global EY organization.