Ask ten Chief Data Officers what their job actually is, and you’ll get ten different answers. Ask an org chart, and you’ll get an even less helpful one. It’s the single most important thing to understand about the role. There really is no standard.
On paper, the definition is straightforward. Gartner describes the chief data and analytics officer as the executive with primary enterprise accountability for creating value from an organisation’s data and analytics assets and ecosystem, with “chief data officer” as the most common equivalent title. In practice, that one sentence is doing an enormous amount of quiet work, because “value creation” can mean governance and risk mitigation in one organisation and revenue generation and AI strategy in another, often within the same industry, sometimes within the same building.
The CDO role is barely a teenager. It emerged out of financial services in the early 2010s, largely as a response to regulatory pressure — someone needed to own data quality, lineage, and compliance after the 2008 crisis exposed how badly banks understood their own data. When NewVantage Partners first ran its executive data leadership survey in 2012, only around 12% of large organisations had appointed a CDO or equivalent, according to reporting on the survey’s history. By 2023, that figure had climbed to roughly 84% of the Fortune 1000 firms surveyed.
But adoption isn’t the same as clarity. The MIT CDOIQ Symposium’s own retrospective on the role notes that the first generation of CDOs was almost entirely occupied with data management fundamentals — quality, access, preparation. Today, that same research finds the majority of CDOs describe themselves as focused on “offensive” work: revenue generation, market expansion, product innovation. The job title stayed the same. The job did not.
When you hire a CDO, you’re not hiring a fixed function the way you would a CFO or General Counsel. You’re hiring for whatever your organisation currently believes data leadership should mean — and that belief is still being formed, industry by industry, sometimes quarter by quarter. The strategy is fluid.
Most CDO mandates sit somewhere across five overlapping areas: data governance and quality, data infrastructure and architecture, analytics and insight generation, data culture and literacy, and — increasingly — AI readiness and strategy. Few CDOs own all five with equal weight, and the balance tells you a lot about why the role was created in that specific organisation.
A CDO hired into a heavily regulated bank is likely still doing serious governance work — lineage, privacy, model risk. A CDO hired into a retailer or media company is far more likely to be judged on monetisation: better pricing, reduced churn, faster product cycles. Neither is doing the role “wrong.” They’re doing different jobs that happen to share a title, which is precisely why comparing CDOs across companies is a much shakier exercise than comparing CFOs or CIOs.
A lot of organisations appoint a CDO before they’ve decided which of these five jobs they actually want done. It’s arguably the leading cause of the role’s problem. They’re often hired under the incorrect mandate where the CDO thinks they’re being brought in to do X, and the company thinks they should be doing Y. It’s no wonder the tenure is so low – more on that in a moment.
Where a CDO sits on the org chart is one of the more revealing signals in the whole conversation. Davenport & Bean’s 2023 survey found in 2023 43.3% of CDOs reporting to the CEO or COO, with 27.4% still reporting into the CIO — a legacy of the role’s early, IT-adjacent origins. By 2025, Gartner’s research showed CDOs reporting directly to the CEO had risen to 36%, up from 21% the year before, a shift attributed to AI elevating the role’s visibility at board level.
That trend line matters because reporting into the CIO tends to frame data as an IT asset to be managed, while reporting into the CEO frames it as a commercial asset to be exploited. A CDO’s actual influence, their ability to reallocate budget, set enterprise-wide standards, or say no to a business unit, is shaped far more by that reporting line than by anything in their job description.
The average CDO tenure has hovered between 2 and 2.5 years for most of the role’s history — MIT Sloan puts it at around 30 months, against nearly seven years for a typical CEO and roughly four and a half for a CFO or CIO. Recent benchmarking suggests a slight improvement, with the proportion of CDOs staying beyond five years rising from around 17% to nearly 26% year on year, which is genuine progress, but it’s progress from a low base, and it hasn’t fundamentally changed the pattern.
Why so short? A few reasons come up repeatedly in the research and in conversations across the executive search world. Organisations frequently hire an outsider as a “change agent,” which builds urgency into the mandate from day one but rarely builds patience into the board’s expectations. Success can be self-limiting too: if a CDO fixes the obvious problem fast, there’s sometimes a sense there’s “nothing left to do” – even though the more interesting, harder-to-measure work of embedding data into decision-making is only just starting. And a genuinely uncomfortable finding from Davenport & Bean’s research is that only around a quarter of organisations report having actually become “data-driven” despite years of CDO investment, which leaves plenty of room for boards to conclude the role — or the individual in it — isn’t working, even when the underlying obstacles are organisational rather than personal.
Just as the CDO role was starting to mature, generative AI reopened the question of who owns what. Gartner’s own analysts have argued that CDAOs who fail to demonstrate company-wide influence and measurable business impact risk having their responsibilities absorbed back into IT — while, somewhat contradictorily, 70% of CDAOs surveyed already hold primary responsibility for AI strategy. Some organisations are now carving out a separate Chief AI Officer position entirely, creating a fresh turf question between CDO, CIO, and CAIO that very few companies have resolved cleanly. Gartner’s own guidance is that this collaboration has to be genuinely seamless to work — which is analyst-speak for “this is going to be messy for a while.”
A senior executive accountable for turning an organisation’s data into something the business can actually use, whether that’s cleaner risk management, sharper decision-making, or new revenue, with a mandate, reporting line, and lifespan that vary enormously depending on how mature the organisation’s relationship with data already is.
The role isn’t poorly defined because the people doing it lack clarity. It’s poorly defined because it’s still relatively new, still absorbing whatever the business’s most urgent data problem happens to be this year, and now navigating an AI moment that’s redrawing the boundaries of the job in real time. The organisations getting the most out of the role tend to be the ones who resist the instinct to write a generic CDO job description and instead ask a much more specific question first: what, precisely, do we need this person to fix — and are we prepared to give them the authority, time, and reporting line to actually do it?
Get that question right, and the title starts to matter a lot less than the mandate behind it.
