CDO challenges that hinder data-driven initiatives
Chief data officers must turn AI ambition into measurable value. These eight issues show where data strategies break down -- and how leaders can respond.
AI spending has outpaced results, putting data leaders under pressure to deliver some measurable business value.
Deloitte's 2026 CDAO survey found that 56% of chief data and analytics officers feel intense pressure to prove the business value and direct ROI of data and AI initiatives. Multiple factors contribute to the low success rate, particularly data quality issues, lack of literacy skills and legacy systems blocking data flow. But CDOs can overcome such obstacles to ensure their -- and their organizations' -- success.
"CDOs are navigating one of the toughest balancing acts in the enterprise," said Sarah Iannantuono, a data, security and risk executive and member of the Emerging Trends Working Group at ISACA. "As the AI conversion gives way to ROI conversations, the focus is once again shifting back to the core requirements of the role: strong data foundations, modernizing legacy systems, clear data literacy training and delivering measurable outcomes."
Data leaders point to eight major challenges hindering data-driven initiatives. Addressing these obstacles requires CDOs to connect data investments to business priorities, governance and defined metrics.
1. Unclear AI objectives
Urged by their boards, CEOs and other executives raced to implement AI across operations, fearing that they'd fall behind their competitors.
"People were experimenting with AI with a sense of urgency," said Carl Gerber, a former Big Four CDO now serving as a strategic advisor and longtime senior fellow with the CDOIQ Society.
However, organizations that rushed implementation often lacked strategies and targets to define success, Gerber said. He noted that only a sliver of companies had a clear vision of where they wanted to go, and those are the ones seeing quantifiable returns. Gerber pointed to a 2025 MIT study that found only 5% of organizations saw value from their generative AI investments. The rest of the initiatives -- the ones without payoff -- tended to be based on a "do something with AI and then figure it out" mindset, according to Gerber.
That approach left CDOs trying to mature their data programs without a defined business outcome or prioritization model. The C-suite needs a clear strategy for their organizations, said Bharath Thota, partner at consulting firm Kearney.
"The top companies start top-down, [establishing] what are the actions that matter, what order to take those actions, what do we need, what data do we need, what can we access, and what can we start with to start getting value," Thota said. "Then the CDO can work in conjunction with the business on data needs and priorities."
2. Big goals but little authority
Even with clear objectives, CDOs often lack the necessary authority to mandate action, said Deepak Seth, senior director analyst with Gartner. He pointed to Gartner research showing less than half of data leaders are given the authority to make necessary growth happen. In those cases, CDOs rely on other executives, such as CIOs, to authorize action. As such, CDOs must rely more on building partnerships, Seth said.
"The smart ones will be almost like political players," he added.
3. Low data literacy
Enterprise data volumes have increased rapidly over the past 10 years. Organizations understand they must find ways to manage and use this data, from adopting business intelligence platforms to advanced analytics tools and AI capabilities. While these technologies are important to the data process, human intervention is still necessary. But most CDOs still find that data literacy is lacking among their workers.
Informatica's CDO Insights 2026 report found that 75% of data leaders believe their employees need upskilling in data literacy, requiring more investments in more education and training.
4. Lack of trust
Employees and executives might use data and AI outputs, but they often hesitate to rely on them for decisions, Thota said.
Recent research shows why confidence can be fragile. The Informatica survey found that 65% of data leaders believe that most or nearly all their employees trust the data they're using for AI. But a 2025 Udacity survey found that 3 out of 4 workers frequently abandon AI tools mid-task, usually because outputs lack accuracy or quality. The same survey found that 45% don't trust a colleague's work if they know AI helped produce it.
"One false outcome can create a bad impression on all the work the CDO is doing on the back end," Thota added.
To build confidence, CDOs must work with other executives to improve data literacy and, thus, trust in the information employees use, Thota explained. That includes making business leaders jointly accountable for data completeness and accuracy.
"People won't trust data until you invest in data culture and data stewardship," said Mark Johnson, the chief growth officer and executive oversight leader for data and analytics at CoStrategix.
5. Data silos and poor data quality
Per Informatica, 57% of leaders point to data reliability as a key barrier to moving AI projects from pilots to production. The 2025 IBM Institute for Business Value report found that 43% of COOs listed data quality issues as their most significant data priority.
There's one cause behind both issues: the data isn't ready. Gerber said many organizations struggle with the data readiness required for AI and noted that many business analysts still turn to data scientists to prepare data. The 2026 State of Data Integrity and AI Readiness report from Lebow College at Drexel University found that 43% of respondents cited data readiness as the most significant barrier to aligning AI with business objectives.
"Companies have invested millions of dollars in their enterprise data strategies, but they still haven't solved their data silo problem," he added.
Experts advise organizations to centralize data into a unified platform with strict governance protocols and strong master data management -- work that requires executive support, funding and cultural changes.
6. Governance and compliance shortcomings
Governance policies only matter if organizations can enforce them. The Drexel survey found that 39% of respondents cited governance obstacles, reflecting the challenge for many CDOs: making policies enforceable.
"It's not just about writing rules and regulations and policies," Thota said. "Governance is about an operating model."
Without that model, organizations can end up with poorly governed data that undermines AI and analytics outputs and weakens trust in the data program, he said.
7. Data and AI skills gaps
According to the Drexel University survey, 41% cited skills as a top challenge.
"Competition for strong data talent can be seen at all levels, with AI and advanced analytic skills increasing demand for talent," Iannantuono said. She also noted that CDOs frequently compete for skilled workers -- and pushing up pay and benefits in the process -- which causes friction between immediate and long-term talent requirements.
Data leaders said competition for data professionals is just part of the talent gap. Organizations also need to train workers in the data skills required for the AI era. But CDOs can't solve this alone, said Iannantuono.
"A concentrated collaborative effort across the business is required," she said. "Organizations that make the most progress within these challenges have a clear view on their current environment -- people, process, technology -- alongside tangible, clear goals and strong executive sponsorship."
8. Budget constraints
CDOs face expanding data environments and growing demands for tangible results from their data programs, but this requires funds to match the ambition.
"That's one of the big challenges that CDOs have, being able to curate the data landscape with adequate resourcing and tooling, which takes budget to orchestrate the mission within the time horizon that the expectation sits within," Johnson said.
CDOs who match up business objectives with data programs can prioritize initiatives with the greatest impact to direct their resources most effectively, Johnson and other experts explained. That alignment also helps them make the ROI case for additional funding, staffing and tools.
Mary K. Pratt is an award-winning freelance journalist specializing in enterprise IT, cybersecurity strategy and data management.
