Home » Technology » AI Forces Many Firms to Rethink Data-Crunching Methods

AI Forces Many Firms to Rethink Data-Crunching Methods

The survey results published Monday by the Association for Computing Machinery (ACM) show that artificial intelligence is forcing companies to evolve, rethinking how they govern and analyze their data. Gartner.

Based on a survey of 479 chief data and analytics officers, chief data officers, and chief analytics officers worldwide, the results revealed that 61% of organizations are reevaluating their data and analytics (D&A) operating models because of the impact of disruptive AI technologies.

When asked about changes chief data analytics officers (CDAOs) need to make to their D&A operating model to be fit for current and future purposes, Gartner found that 38% said their D&A architecture will be overhauled over the next 12 to 18 months.

29% of respondents said that they will also adopt governance policies, standards, and practices and revamp their management of data assets.

“Responding to the rapid evolution of D&A and AI technologies, CDAOs are wasting no time in making changes to their operating model,” Gartner Vice President Analyst Alan D. Duncan said in a statement.

“While the management of their organization’s D&A operating model is increasing year over year, no other role than the CDAO has the responsibility of many of the key enablers of AI, which include data governance, D&A ethics, and data and AI literacy,” Duncan continued.

“The CDAO’s role has grown in scope as budgetary and resource constraints are becoming more difficult,” he said.

Analogy: AI Musical Star

The survey results, released in advance of Gartner’s Data & Analytics Summit to be held in London in May, noted that AI has become part of the duties of a majority of CDAOs, with 58% saying AI is within their scope of responsibilities, an increase from 34% in 2023.

Donna Medeiros, Gartner Senior Analyst, said that many organizations, particularly in the private sector are redefining their business models based on AI. This is either for automation, operational excellence, or to forge new business paths.

She told TechNewsWorld that “there are many industries who have used AI for years, but gen AI in particular has undergone a number of pilots.”

Forrester also stated in their report “AI Ready For The Light, But Data And Analytical Determines If It Shines”, that “Outputs from AI technology will only be as good the data that went into them. Business and Technology Leaders must embrace best practice for data and analytical to maximize the value of their artificial intelligence.”

“If generative AI is the star actor in ‘Business Technology: The Musical, 2024,’ data and analytics leaders are the stage managers,” noted the report, authored by Forrester analyst Zeid Khater with Aaron Katz, Kim Herrington, Jayesh Chaurasia, Brandon Purcell, Noel Yuhanna, Evan Megan, and Jen Barton.

Set yourself up for success

Forrester said that before the star technology is able to deliver value to its stakeholders and customers, leaders in data and analytics must set up the right people, processes and platforms.

Data platforms and data-quality practices will be used to support “forward-thinking” data and analytics teams. [large language models] Khater wrote that “AI technologies will be able to provide value to organizations if they have the right data, data skills and data unstructured data.”

He added that “the work done by data and analytics teams behind the scenes will determine if new AI technologies fail or succeed when they are brought to the forefront.”

Forrester stressed that improved data quality allows machine-learning models to accurately identify patterns and improve prediction accuracy.

Khater said that organizations who prioritize data quality by using rigorous metrics, assessments and continuous monitoring can not only improve the model’s reliability, but also reduce any risk of bias or erroneous results. This improvement in model accuracy results in cost savings, better business outcomes, as well as a competitive edge.

Need for Metrics

Gartner has found that CDAOs have lagged behind in terms of metrics. It noted that only 49% of surveyed CDAOs have established business outcome-driven metrics that allow stakeholders to track D&A value. In addition, 34% have not established business outcome metrics for D&A.

“Forty-nine percent is low because it means that less than half understand that they need to work with the business to put metrics in place to demonstrate the value of what they’re accomplishing because invariably they’re going to be asked, ‘What did we get for this investment?'” Medeiros said.

“The D&A leader can’t do this in a vacuum,” she continued. They will know what is important to their business and what they need to do. [key performance indicators] “They can then get to the real metrics.”

She explained that “CDAOs must work with all business units to estimate and project the impact of return on investment for all use-cases based on metrics set up during planning to understand what they achieve.” “D&A leaders need to engage with business folks to establish metrics and get well above the 49%.”

She continued, “CDAOs should also feel comfortable discussing business discipline.” “That’s not easy for many because they haven’t been in the D&A leader role for very long. Most people are still relatively new to the role.

Only 22% of CDAOs, according to the survey have held their positions for at least five years. 56% of CDAOs surveyed said they are looking for a new job. Medeiros explained that this is because the position has so many demands.

Power Play is Needed

Gartner found that CDAOs face significant costs associated with expanding their responsibilities. CDAOs reporting an increase year-over-year in the funding of their function still have budget constraints.

“CDAOs who present better business cases to CFOs receive better and quicker funding for their D&A initiatives. Duncan stated that “they also gain greater executive buy-in”.

Gartner stated that CDAOs must increase their power and influence in order to achieve results. To demonstrate their value, they must also understand the organization’s value levers and problems from beginning to end.

Duncan predicted that by 2026 75% of CDAOs will have been assimilated to technology functions if they fail to prioritize organization-wide impact and measurable impacts.