CDAO vs CTO vs CIO: Who Actually Owns Data at Your Company?

At every company past a certain size, there is a data ownership vacuum. I have dealt with it repeatedly — at Adidas running a $5B platform with 500+ engineers, at Sweetgreen as combined CTO/CIO, and at Bain Capital portfolio companies where the question "who owns the data strategy?" reliably produced three different answers from three different executives.

The CDAO role — Chief Data & Analytics Officer — exists to fill that vacuum. But before I explain what a CDAO actually does and how they differ from a CTO and CIO, I need to clear up a title confusion that trips up almost every conversation about this role.

The Data Ownership Vacuum

Here is the pattern I have seen at every company past roughly $100M in revenue. Engineering — under the CTO — builds the data infrastructure: pipelines, data warehouse, data lake. They do this work because it is technically interesting and because the product needs it. Separately, the business builds dashboards and reports using BI tools. They do this because they need to measure things. Nobody bridges them.

The result is a vacuum. Engineering owns the pipes but does not know what the business is trying to measure. The business uses the data but cannot tell you which table is authoritative or whether the numbers have been cleaned. Both sides are operating in good faith, but the gap between them produces the most familiar failure mode in analytics: nobody trusts the data.

The CDAO emerged to own that middle ground. Not just the infrastructure (that is CTO territory) and not just the business intelligence outputs (that is BI leadership territory) — but the strategy and governance that connects them.

The CDO to CDAO Evolution

CDO — Chief Data Officer — was the 2010s title for this role. Companies created it primarily to address data governance: who defines "active customer"? Who is responsible for data quality? Who owns the data compliance obligations under regulations like GDPR and CCPA?

CDAO emerged as companies realized that separating data engineering from analytics was a mistake. In the CDO model, you often had a data engineering team owned by the CTO and an analytics or BI team owned by the CDO (or scattered across the business). These teams should be integrated — analytics engineers need to work directly with the data engineers who build the pipelines. When they do not, you get the 400-table warehouse problem.

The modern CDAO title explicitly combines both: governance AND analytics. One leader owns the full data value chain from infrastructure to insight.

What a CDAO Actually Owns

The CDAO mandate covers more ground than either the CTO or CIO deal with on the data side. In concrete terms, a well-scoped CDAO owns:

• Data strategy: what data assets to build, what to monetize, and the roadmap for building data as a competitive moat.
• Data governance: definitions (who decides what "active customer" means?), quality standards, data ownership across the organization, and the processes that keep it all coherent over time.
• Data platform and infrastructure: the warehouse, data lake, and pipelines — often shared responsibility with the CTO, but with the CDAO setting the roadmap.
• Analytics and BI: dashboards, self-serve analytics, reporting tooling, and the analytics engineering function that sits between raw data and business consumption.
• Data science and ML feature engineering: at many companies this sits under the CDAO because data scientists depend on clean, governed data more than any other function.
• Regulatory data compliance: GDPR, CCPA, HIPAA where applicable — data residency, right to deletion, consent management. Not corporate security broadly (that is the CIO), but the data-specific compliance obligations.
• Data products: internal tools, external data monetization, and treating data as a product rather than a byproduct.

CDAO vs CTO: The Platform Question

The clearest distinction: the CTO builds the data infrastructure; the CDAO governs and monetizes what flows through it. But the line is not as clean as it sounds, because the most contested territory is the data platform team itself.

The CTO's argument: "Data infrastructure is infrastructure. It lives in engineering. Our data platform team is an engineering team. They need to maintain engineering standards, follow engineering processes, and report to engineering leadership."

The CDAO's argument: "Data is the product. The data platform is the foundation of everything I am accountable for. If I do not own the platform roadmap, I cannot be held responsible for the quality or reliability of the data. Put the data platform team in CDAO."

Both positions have merit. My recommendation, after working through this dispute at multiple companies: give the data platform team to the CDAO for roadmap ownership, but have them embedded in engineering reporting to ensure technical standards are maintained. It is a matrix structure, which is never ideal, but it is the least bad option for most organizations. The alternative — a clean org boundary — tends to produce either a data platform that is technically sound but strategically disconnected, or one that is strategically aligned but architecturally unsound.

CDAO vs CIO: Different Questions

The CDAO/CIO relationship is less contentious than CDAO/CTO, because the two roles are asking fundamentally different questions about data.

The CIO cares about data from a security and system-of-record perspective: who has access to what data? Which systems are the source of truth for employee records, financial data, customer data? Where does data live for residency and compliance purposes? The CIO is protecting and auditing data as an operational asset.

The CDAO cares about data as a business asset: what are we measuring, how do we build data products, how do we extract analytics value from the data we have? The CDAO is building and exploiting data as a strategic resource.

These complement each other and rarely conflict directly, with one exception: definitions of system of record. When the CIO says "Workday is the source of truth for headcount" and the CDAO says "our analytics warehouse is the source of truth for reporting headcount," you have a problem. This gets resolved by the data governance function — which sits under the CDAO — establishing clear data lineage and canonical definitions. But it requires explicit agreement with the CIO's team.

CDAO vs CTO vs CIO: The Comparison

When to Hire a CDAO

There are five situations where I think a dedicated CDAO is genuinely justified:

Data is a direct revenue stream

If your company sells data products, uses data-driven pricing at scale, or runs ML-driven personalization as a core business differentiator — not just a feature, but a real competitive moat — then data needs C-suite advocacy. The CDAO ensures that data investment is treated as capital investment in a core business function, not as overhead.

Regulatory data obligations are complex

Financial services, healthcare, and any company operating across multiple jurisdictions with data localization requirements have data compliance obligations that are genuinely full-time executive work. GDPR data subject rights, CCPA opt-out management, HIPAA data handling, cross-border data transfer agreements — the CDAO who also reports to the board on these topics is much better positioned than a VP who is managing this alongside a full analytics mandate.

BI chaos: 50 versions of the same number

If your company has multiple competing dashboards showing different answers to the same business question, you have a data governance problem masquerading as a technology problem. Engineering can build more pipelines; that does not solve the underlying issue of who owns definitions. A CDAO with a proper data governance function fixes this. A VP of Engineering cannot, because governance requires cross-functional authority that engineering leadership does not have.

Building AI/ML products that need a reliable data foundation

ML models are only as good as the data they train on. If your company is building AI products in earnest, the data foundation — feature stores, data labeling pipelines, training data governance — is as important as the model development. CDAO ownership of that foundation gives AI initiatives a reliable substrate. Without it, every AI project starts by rediscovering the same data quality problems.

Large data organization without strategic leadership

If you have 20+ data people — data engineers, analysts, data scientists — spread across the company without a unifying strategic leader, you have a problem waiting to surface. Teams are duplicating work, using inconsistent definitions, and building in different technical directions. That is a CDAO problem.

When the CTO or CIO Can Cover It

Most companies do not need a CDAO yet. The role earns its keep at a particular level of organizational maturity and data complexity. You probably do not need a dedicated CDAO if:

• The company is under 300 people. A VP of Data reporting to the CTO is almost always sufficient at this stage.
• The analytics function is small (fewer than 10 people). It does not need C-suite leadership — it needs solid management and a data-competent engineering organization.
• Data is an operational input rather than a revenue driver. If data is how you run the business but not how you make money, the CTO can govern it adequately with the right VP of Data below them.
• You are pre-product market fit. Do not add executive overhead before you have something to optimize.

The Honest Bottom Line

The CDAO role is not a trophy — it is a solution to a specific organizational problem. That problem is the data ownership vacuum: the gap between engineering building data infrastructure and the business consuming analytics outputs, with nobody bridging them strategically.

If your company has that vacuum and has grown to the point where a VP of Data cannot fix it — because it needs cross-functional authority, board-level reporting, or a direct mandate around data as a business asset — then the CDAO hire is justified and probably overdue.

If you do not yet have that vacuum, or if it is small enough that a strong VP of Data can manage it, hold off. The CDAO title brings overhead: reporting lines to negotiate, scope boundaries to draw, another C-suite member in every strategic conversation. Only add it when the organizational complexity genuinely requires it.

Frequently Asked Questions

What's the difference between CDO and CDAO?

CDO (Chief Data Officer) is the legacy title that emerged in the 2010s. CDAO (Chief Data & Analytics Officer) is the modern evolution: it explicitly adds "Analytics" to reflect ownership of both data infrastructure/governance AND analytics outputs. The practical difference is that CDAO ends the organizational separation between a data engineering team and a BI/analytics team — both report to the same executive. Many companies that still use the CDO title are functionally running a CDAO model.

Does CDAO report to CTO or CEO?

It varies, and the answer matters. At data-mature companies where data is a core business driver or revenue stream, the CDAO reports directly to the CEO — this signals that data is a strategic asset, not an engineering function. At technology companies where data is a capability rather than a product, the CDAO often reports to the CTO. In financial services, CDAO sometimes reports to the CFO because data governance is primarily a regulatory compliance function. Neither is universally right. What matters is that the reporting structure matches the mandate.

What background do CDAOs come from?

CDAO backgrounds are more varied than CTO or CIO backgrounds. Common paths: data science or analytics engineering leaders who grew into the business strategy side; product managers who specialized in data products; data-focused consultants with strategy experience. Increasingly, CDAOs come from ML engineering backgrounds, particularly at AI-first companies. This is quite different from the legacy CDO background, which often came from IT, enterprise data warehousing, or compliance-heavy industries.

What is a typical CDAO salary?

Median US total compensation for a CDAO is $280K–$400K. Financial services pays the highest — $350K–$550K for CDAOs with P&L ownership and regulatory mandates. Technology companies run $260K–$380K. The range is wide because the CDAO role varies significantly in scope: some CDAOs own a 200-person data organization including data platform, data science, and analytics; others are more of a chief data strategist with a smaller team.

When does a company need a CDAO vs a VP of Data?

VP of Data = operational data leadership with CTO oversight. CDAO = C-suite visibility, board-level reporting, and data treated as a core strategic asset. The distinction matters at roughly $500M revenue and above, or earlier if data is a direct revenue driver. At a company that sells data products, monetizes its dataset, or runs ML-driven personalization at scale, the CDAO title (and the board access that comes with it) is justified much earlier.

Is CDAO the same as Chief Analytics Officer (CAO)?

No. The Chief Analytics Officer typically owns analytics and BI only — dashboards, reporting, business intelligence. CDAO is broader: it adds data engineering, data governance, data platform ownership, and often data science and ML. A CAO answers the question "what are the numbers showing us?" A CDAO also asks "do we trust these numbers, where do they come from, and how do we build data as a business asset?"

What is the relationship between CDAO and CAIO?

CDAO owns the data that trains AI models — the raw material. CAIO owns the models and their deployment — the product built from that material. They must work closely or the relationship breaks: data quality bottlenecks will block AI initiatives, and AI demand will shape data collection priorities. The CDAO-CAIO relationship is one of the most important partnerships to get right as companies build AI products at scale. At some companies, these roles are combined; at others, they are deliberately kept separate to maintain a check on data quality.

Can one person be CTO and CDAO?

Yes, and it is common at companies under 300–400 people. A data-competent CTO often covers both — building the infrastructure and thinking strategically about data governance and analytics. The CDAO role makes sense as a separate function when the data organization is large enough (typically 15–20+ people) to need dedicated C-level leadership, and when data is strategic enough that it needs its own board-level advocate.