Banks’ climate pledges don’t add up to change, research finds

The impact of misleading headlines on Facebook

Meet the 6 new faculty members joining MIT Sloan in 2024

Credit: Mimi Phan / Olga Shestakova / iStock

Ideas Made to Matter


Chief data officers don’t stay in their roles long. Here’s why


There’s clear evidence that enterprises around the world are embracing data and analytics. The 2022 senior data and AI leadership survey from NewVantage Partners found that 74% of firms have a chief data officer or chief data and analytics officer, while 92% of firms are seeing returns from data and AI investments.

But at the same time, leaders in charge of these initiatives aren’t sticking around. Additional research suggests that the average tenure of a CDO is just 2 1/2 years. This compares to nearly seven years for the typical chief executive officer and just over 4 1/2 years for the average chief financial officer or chief information officer.

There are many reasons a CDO may decide to depart, but a main one is often a failure to articulate from the very start how a CDO’s work differs from that of other executives.

“When you carve off a CDO or [chief data analytics officer] role for the first time, you’re carving off an area of responsibility from someone else. In many cases, it’s the CIO. Building that relationship and agreeing upon areas of responsibility between IT and the CDAO organization is vitally important,” said Ajai Sehgal, CDAO at the Mayo Clinic. “If not, then you’re going to have an adversary for the rest of your tenure.”

At the recent MIT CDOIQ Symposium, Sehgal and other data officers discussed how CDOs can deliver business value in a short time and extend their tenures if they want to stay longer.

Understand that progress can be slow

According to the NewVantage survey, which included 94 executives from Fortune 1000 companies, nearly three in four companies have a CDO in place. But individuals in those roles find progress to be slow:

  • Only half of CDOs are able to drive innovation using data.
  • Just 40% of CDOs manage data as a business asset.
  • Roughly 26% of CDOs have succeeded in creating a data-driven organization.
  • About 25% of CDOs have no single point of accountability for data within their organization.
  • Only 40% of respondents said the CDO role is successful and established within their organization.

Two factors further stifle progress. One is that it’s common for organizations to hire an outsider for the CDO role — someone regarded as a “change agent,” NewVantage CEO and founder Randy Bean said. An executive brought in to lead transformative initiatives or otherwise shake things up may find it difficult to get along with leaders who have risen through the internal ranks.


In a recent survey, 40% of executives said the CDO role is successful and established within their organization.

The other factor is the short tenure of most CDOs. (None of the four panelists had been in their role longer than 18 months, and one — Best Buy CDAO Craig Brabec — was in his second week on the job.)

In a 2021 Harvard Business Review article, Bean and co-author Tom Davenport of the MIT Initiative on the Digital Economy suggested that the honeymoon period for the CDO lasts roughly 18 months — at which point it’s expected that they have achieved “major transformational change.” If that has happened, then a CDO may be motivated to move on to another challenge at another company. If it hasn’t, then the pressure to get results quickly could push a CDO out the door.

Make sure the CDO and the business understand each other

Panelists pointed to a host of challenges facing the CDO role. Most stem from corporate culture, not the limitations of technology.

“Many organizations just jump at the excitement of announcing a new chief data officer, but there aren’t clear expectations or the desired delivery of value to be created from that role,” said DeWayne Griffin, vice president and CDO at State Farm. “You can get caught in a five-year success journey, or you can get pulled into delivering value in the next 12 months.”

Brabec and Sehgal noted the importance of aligning expectations to specific business outcomes. This can be difficult for leaders who traditionally work as technologists and may not know business operations firsthand.

To close this gap, members of Sehgal’s technology team are paired with Mayo Clinic physicians to learn about the many complexities of healthcare delivery. In his prior role as CDAO at McDonald’s, Brabec assigned data scientists to shifts in a restaurant.

“To be successful with the models they were building, they needed to understand how what they were doing impacted operations,” Brabec said.

While CDOs need to understand the business, the business needs to understand what CDOs and their teams are there to do.

Grace Lee, senior vice president and CDAO at Scotiabank, said many organizations need to evolve their mindset when it comes to data and analytics.

“Data and analytics isn’t a side project. Until we understand that the entire business runs on data, we are going to be consistently challenged with and limited by the project-based mindset around the CDO role,” she said. “As a CDO, you need to know that you’re an equal partner in driving the business forward, not just a hired gun that’s been brought in for your technical expertise.”

Show the business what’s possible with data

To help the business understand their value, panelists described the CDO role serving as a data evangelist: Showing individual business units what’s possible when previously untapped data resources become core parts of their everyday work.

“Just as the CFO is responsible for the flow of money through an organization, the CDO job is equally foundational, responsible for creating the free flow of information across the company, making sure that it's accessible, it's safe, it's used ethically, and it's well protected,” Scotiabank’s Lee said.

For Scotiabank, this free flow of information translates to improving the efficiency and effectiveness of the bank’s operations and an improved customer experience. For the Mayo Clinic, this means making sure that doctors have access to in-depth analytics and decision support that doesn’t interfere with their conversation with a patient, Sehgal said.

State Farm is taking a more literal approach, Griffin said, by specifically measuring how the flow of data leads to better business decisions.

“One of the outcomes we’ve rallied around is the life cycle of data. What's the time it takes from acquisition from requests to actual usage?” he said. “We’re actually measuring the accessibility and usage of data through the organization.”

Understand that success – and tenure – look different for everyone

Better access to analysis, fewer cancelled jobs, and a shorter data life cycle all represent wins for a CDO. But are they enough to convince someone to stay? Not everyone agreed.

The job can be stressful, and some CDOs will burn out in 2 1/2 years.

“You need to have a long-term strategic plan with a vision that you can evangelize — but not at the expense of having a tactical plan that allows you to execute and deliver business value now. You need both,” Sehgal said.  

Related Articles

How to build a data analytics dream team
CDOs look to the future
How to build a data-driven company

Many CDOs leave on their own accord, often looking for opportunities to create more impact. Few retire or get fired, but many find better opportunities elsewhere. (Among the four CDO panelists, only State Farm’s Griffin was an internal hire.)

 “It can be difficult to effect change and feel like you’re having an impact, and it can be a lonely job if you have misaligned expectations,” Lee said.

Plus, if a CDO achieves rapid results, there may be little else to accomplish. Brabec’s tenure at McDonald’s lasted less than two years, but in that time McDonald’s saw a 60% increase in sales from digital channels in its largest markets.

While every CDO role is different, Griffin said effective relationships with business partners should make anyone’s job easier.

“If you forge the right relationships, then you can identify the right problems and set the right expectations for what success looks like,” Griffin said. “It’s founded on relationships, on outcomes, and on rallying the organization around the right pieces.”

Read next:  Making the business case for a chief data officer

For more info Sara Brown Senior News Editor and Writer