Getty Images
What a leadership transition reveals about your data strategy
The failure pattern of a leadership transition reveals what data strategy institutionalization requires. Without it, a data strategy framework is a dependency, not a strategy.
When a CDO departs, organizations often discover that their data strategy was little more than the vision of a single executive, never embedded in the organization's culture and fleeting in its efficacy.
When that's the case, vendor relationships strain, governance initiatives stall and undocumented processes that existed only in the CDO's mental roadmap surface as gaps -- a pattern that plays out before, during and after a departure. Any data strategy centered on a single executive is an enterprise liability, and the average CDO tenure of 2.5 years makes that liability a recurring one.
The early warning signs
The warning signs that a data strategy is tied to one executive appear well before a departure. Most stem from a lack of formalized governance and ownership. According to Dominic Sartorio, vice president of product marketing at Denodo, CDOs often "come in with a great vision, but the issue is they lack the ability to change-manage the rest of the organization. They lack the ability to operationalize. After they leave, the vision goes with them and falls apart."
Organizations also lack quantifiable measures to determine whether the data strategy is working, such as dashboards that track data usage and its impact on business goals. "The early warning signs [include] a lack of data contracts that are agreed to by both sides," Sartorio said. "There is a lack of identified business users who take some ownership stake that the data is supporting the outcome."
Governance initiatives that remain one executive's priority rather than a shared responsibility, have no designated data governance stakeholders "in terms of data owners, data stewards and data custodians," said Thomas Phelps, Laserfiche CIO and SVP of corporate strategy. Litigation, regulatory penalties and data breaches follow when governance, data access and security lack clear ownership.
What a departure exposes
When CDOs depart an organization, it becomes apparent if their data strategies were attempts to operationalize their visions. During knowledge transfer sessions, significant gaps emerge between how the data strategy was supposed to work and how it actually does. Those disparities include what Swaroop Jagadish, CEO of DataHub, called "the work around fragile dependencies; the institutional judgments that got made along the way and never got written down."
Much of the CDO's data strategy never leaves slide decks and meetings. "Every CDO carries an internal map of which business use case to prioritize, which data is trustworthy, where the fragility is, and why certain decisions were made," Jagadish observed. "All of that knowledge can disappear. It's too focused on an individual."
What breaks after departure
After CDOs leave an organization, points of failure can include strained vendor relationships and a lack of production in data-driven processes. Without the personal knowledge of which data pipelines to prioritize and how to optimize them, operational output drops. Pega CTO Don Schuerman said, "end user customers are falling off of the product usage cycle." Moreover, vendor relationships built on the CDO's personal trust fray, reinforcing that "your data strategy is not to implement Snowflake; it's to surface insights," Schuerman said.
For data governance initiatives, there may be a lack of ownership throughout the organization when data breaches or regulatory compliance penalties are incurred. All of these woes are evidence that the data strategy designed to support an organization never actually left the purview of the individual championing it. Such a shortfall is a multifaceted liability. "Three things walk out the door when CDOs leave," Jagadish explained. "First, the vendor relationships built on personal trust. Second is the initiatives: CDOs in particular have talked about governance programs because it's their priority, but they were not institutionalized. The third thing is the mental model of the data landscape."
Building a data strategy framework that lasts
Organizations can avoid the woes that typically accompany a CDO's departure by institutionalizing a data strategy that can outlast the executive's tenure. "A data strategy is too big to be one person's job," Schuerman said. "It demands that all of the leaders who would be consumers of the data strategy have enough of a hand in and understanding of it, so they're not dependent on one person to execute it." That means aligning the strategy with the business processes it supports. Phelps said that alignment is essential to tackling specific use cases for the business.
Documentation is the other half of institutionalization. Strategies formalized through content repositories, knowledge management platforms, data catalogs, vector databases, NLI interfaces and data marketplaces survive executive turnover. Sartorio's point on the necessity of data contracts is part of the broader data product framework. Documentation includes "governance and security policies tied to it, like what the quality target should be, whether it is sensitive data or not, and latency considerations," Sartorio explained. "These are the SLAs and policies a platform can serve up."
Jelani Harper is a data industry analyst and journalist covering data management, AI and enterprise IT for more than a decade. He is a research lead at Blue Badge Insights, writing analyst reports for GigaOm and articles for VentureBeat and The New Stack.
