Execs admit to making ‘material’ decisions based on bad data

Dive Brief:

Nearly half (47%) of finance and technology executives have made a “material” business decision during the past year based on inaccurate, incomplete or outdated financial data, according to a recent survey by enterprise software provider OneStream.
Over 70% of respondents said bad data had cost their organizations at least $500,000, while 37% reported losses exceeding $1 million. The findings point to mounting pressure on organizations to tighten their data governance efforts as artificial intelligence deployments expand across finance operations, according to OneStream.
“Unless companies have data they can trust, AI will only accelerate and amplify bad decisions,” Tom Shea, CEO of OneStream, said in the announcement.

Dive Insight:

The research comes as Gartner projects worldwide AI spending will surge to $2.5 trillion in 2026, underscoring how rapidly companies are scaling investment in the technology.

Nearly 45% of CFOs are feeling pressure from leadership to “do something with AI,” with about six in 10 rolling out AI agents just to see what they can do, according to separate survey results reported in February by finance software firm Basware.

A majority of finance leaders “admit they are largely experimenting with the technology and flying blind when it comes to putting it into practice and delivering ROI,” Basware said in a press release at the time.

Younger executives (ages 25 to 44) are heavier AI adopters, with 82% reporting use of three or more AI tools for decision-making, compared to 69% of more experienced peers, OneStream found. They are also more exposed to risk: over half (51%) report making a material decision based on faulty data, versus 39% of older leaders.

Roughly 80% of leaders say their data governance frameworks are strong enough to support large-scale AI adoption, and 85% report having a formal governance program either in place or underway, according to OneStream’s report. Yet nearly two-thirds (61%) of respondents said they second-guess their data at least monthly.

“Every generation of finance leaders has had to earn trust in the numbers before they could act on them,” Pam McIntyre, chief accounting officer at OneStream, said in the release. “What's different today is that AI has compressed that timeline dramatically, and the margin for error has shrunk with it.”

OneStream surveyed more than 350 senior finance and decision-makers across the U.S., U.K. and France.