How the CDO role is changing in 2026

The CDO role has outgrown the job description most organisations wrote for it. Most CDOs already understand this, but what's harder to sit with is the structural problem the change has created: the new mandate cannot be delivered from inside the data function.

CDOs are now being judged on decision quality across the enterprise, on the same financial measures that govern the rest of the C-suite. The decisions producing those outcomes are made in operations, finance, supply chain, and sales, often by people who in most organisations have not been equipped to use data and AI well in the flow of their work. 

The accountability has moved, but the capability to deliver against it has not.

The shift in brief: CDOs are now answerable for EBITDA, growth, and risk reduction outcomes that depend on workforce-wide decision-making. Most enterprises have invested heavily in platforms and governance, but not in the data and AI fluency of the people whose decisions now sit inside the CDO's accountability. Closing that gap is the strategic question for 2026.

This was the thread running through the DataIQ 100 Summit North America in Nashville this April, where Data & AI Literacy Academy spent two days listening to the most senior data and AI leaders in US enterprise describe a job that has moved well beyond the function it was built around.

"The era of the dashboard CDO is over," said Sushma Punuru, Global Chief Data Officer at Global Payments. "The most successful data and AI leaders will stop being passive stewards and become chief decision and value officers."

The four shifts redefining the CDO role

DataIQ's research at the summit identified four dimensions of the change:

• From enabling analytics to owning decision outcomes
• From experimentation to operational delivery at scale
• From technical expertise to leading organisational change
• From functional leadership to enterprise operator

What links them is exposure. CDOs are now answerable for decisions made well beyond the function they run.

"The question leaders will be judged on will no longer be 'How advanced is your AI?' but 'Which decisions are materially better because of it?'" said Sandeep Mahajan, Director of Data Science at Walmart.

Deepak Jose, Vice President of Data and Decision Intelligence at Niagara Bottling, framed the same shift in financial terms. "Boards and CFOs will increasingly judge data and AI leaders on EBITDA impact, growth, and risk reduction rather than platforms, models or use cases delivered."

From pilots to production

Most large enterprises have proven what is possible in controlled conditions. Translating that into delivery at scale is where the work gets harder, and where most are still stuck.

Manish Agarwal, Vice President of Data and Analytics at Skechers, located the shift specifically in execution. Automation and intelligent agents, he said, will drive the move from enablement to scaled execution and direct value creation. The point that mattered was the next one: what counts is integrating those systems into real business workflows, not running pilots.

The difference is operationally significant. A pilot is contained: variables can be controlled, users can be selected and outputs can be reviewed. 

Scaled execution is none of those things. 

It runs across functions, across decision-makers, across people who were not part of designing the system. The capability the system needs from its users changes accordingly.

Meaghan Ferrigno, Senior Vice President, CFO and CDAO at Destination Canada, named what closing that gap actually requires. Embedding data and AI into daily decision-making, she said, demands sustained cultural change, not just technical progress. Success will be measured by the business models, opportunities, and competitive advantages that data and AI enable, not by what is built.

That matters because most investment to date has gone into building. Platforms, models, infrastructure, governance. The next phase of return depends on what people across the organisation do with those systems once they meet them in the workflow. Which is a workforce question, not a technology one.

Where the new mandate exposes the gap

Investment has gone into platforms, governance, and centres of excellence. Far less has gone into the capability of the workforce that uses what those investments produce.

Tifani McCann, VP of Data and Analytics at Otsuka Pharmaceutical, located this in how the CDO role itself has changed. Technical mastery opens doors, she said, but understanding motivations, perspectives, and relationships shapes your effectiveness as a leader. The job is no longer running a technical function. It is shifting how the rest of the organisation thinks and acts.

Anu Krishnan, Chief Data and Analytics Officer at Chevron, described the same shift in operational terms. Keeping pace with the speed of data and AI, she said, requires empathy and human-centred leadership, building change management programmes that reduce fear and increase adoption.

Fear and adoption are the right words. When AI-supported systems land in a workflow, the people using them are being asked to change how they make decisions, often without being told why or how. The default response is either over-reliance or avoidance. Both produce poor outcomes, and both compound at scale. This is the gap the new mandate exposes. The CDO is now accountable for the quality of decisions across the business, but the people making those decisions sit outside the data function and, in most cases, outside any meaningful programme of capability-building.

If your organisation is wrestling with this gap, our enterprise data and AI fluency programmes are built specifically to close it.

What the data shows

The leaders running these organisations already know where the problem is.

Deloitte's State of AI in the Enterprise 2026, a survey of more than 3,200 business and IT leaders across 24 countries, found that insufficient worker skills are the single biggest barrier to integrating AI into existing workflows. The same study found that 84% of organisations have not yet redesigned jobs or workflows around AI capabilities. The diagnosis is widely shared. The structural response is not.

While that gap sits open, the workforce continues to use AI every day. The KPMG and University of Melbourne global study (Trust, Attitudes and Use of Artificial Intelligence, 2025) found that 58% of US workers rely on AI without properly evaluating the outcomes. 57% have made mistakes in their work as a result.

The mistakes are being made in operations, finance, supply chain, and sales. These are the functions whose decisions now sit inside the CDO's accountability. Most CDOs do not have visibility into where AI is being used across them, let alone confidence in how well it is being used.

What this means for organisations

Aravind Jagannathan, Senior Vice President and Single-Family Chief Data Officer at Freddie Mac, described what the role now requires: embedding AI into products and services, integrating data and AI into business processes, building agile, outcome-driven teams. That is the scope of an enterprise capability, not a function-level project.

It is also a scope that no data function can deliver alone. The CDO is being judged on outcomes produced by people across the business, most of whom sit outside any structured programme of data and AI fluency. The systems can be built centrally, but the decisions cannot.

For organisations taking the new mandate seriously, the work shifts toward the capability of the people whose decisions now sit inside the CDO's accountability. This is the work Data & AI Literacy Academy is built around. Our enterprise data and AI fluency programmes build capability at the points where decisions actually get made, so that the systems data and AI leaders are deploying produce better outcomes in the workflow rather than faster ones. That is how the new mandate gets delivered in practice.

The CDO role has changed. The organisations that close the gap between the new mandate and current workforce capability will be the ones that get the return.

‍