Tech budgets: CFOs battle the bulge
IT costs have become more variable as vendors refine pricing models.
• 4 min read
There’s a good chance your organization’s tech budget is bloating, thanks to spending priorities like cybersecurity and AI making it hard to close the pocketbook. This brings up a dilemma for CFOs: How can companies practice cost discipline without stifling innovation?
Experts we spoke with offered a very human solution to the tech-spending dilemma: teamwork and communication. But that’s easier said than done, they added, because there tend to be walls between the key functions of finance and IT.
“CFOs and CIOs…use a lot of the same terms, and oftentimes mean fundamentally different things,” according to Powell Latimer, market research and analytics manager at the Technology Business Management (TBM) Council. This de facto language barrier can create “two separate spreadsheets that each different side is using for truth,” which is a great way to create perpetual headaches at quarterly business reviews, he told CFO Brew.
Organizations can overcome these barriers by establishing a common language between departments, and getting granular on all the inputs of IT spend—breaking it out, for example, into tech, labor, and cloud costs.
“The people, organizations, and teams that are able to differentiate themselves from the masses are the ones that are going to be able to continually have a conversation about not only the value that they are getting, but the value that they want to get” from technology investments, Latimer said.
Adam Goldbruch, CFO of property-management software developer DoorLoop, knows firsthand the risk that organizations can run into with disorderly AI spending. The company has numerous departments, and “each department wants to use their own AI tool,” he told us. But without collaboration, the company risks creating “a lot of [overlap] in terms of cost.”
The big picture. Recent research suggests that IT spending is likely to keep growing.
Companies spend, on average, nearly 13% of total revenue on IT, according to a recent TBM Council and Information Services Group (ISG) survey of 200+ TBM practitioners. The report also noted that nearly half (47%) of IT costs are now variable, “as more tech vendors move to consumption-based models.”
Additionally, consulting firm West Monroe found last fall that 86% of companies increased IT spending during the prior 12 months, and 85% expected the same for the coming year, the Journal of Accountancy reported.
News built for finance pros
CFO Brew helps finance pros navigate their roles with insights into risk management, compliance, and strategy through our newsletter, virtual events, and digital guides.
By subscribing, you accept our Terms & Privacy Policy.
In a recent LinkedIn post, Justin Etkin, COO of procurement platform Tropic, wrote that almost all of the “fastest -growing” vendors, including Anthropic and Ramp, charged on consumption. This model used to be limited to a small number of vendors, namely cloud platforms like AWS and Snowflake, Etkin told us.
“What we are seeing now is a trend where every software business, both the new AI-native providers and a bunch of the legacy software tools, are transitioning their business models to align with consumption credits outcomes,” he said.
ROI drives. DoorLoop has an internal AI enablement team that both establishes best AI practices and monitors AI initiatives, according to Goldbruch. Employees must show Goldbruch and others on the committee how their AI tool is either saving the company money or increasing revenue. “Every team that wants to use an AI tool will need to present how it’s been ROI-efficient to use this AI tool,” Goldbruch said.
“It all can be measurable,” he added, “and my request is [they] give a milestone and timeline,” such as dollars saved by a certain quarter. “Then we’re going to have some post mortem after that, to see if it makes sense. If not, we can actually stop the project right away, and they will know that they need to put their efficiency there.”
The internal AI team will share successful use cases with other departments they think will see a similar benefit. For example, an AI agent developed by the sales team may also be useful to the customer success team, he said.
Etkin said 2025 was a year of “rapid experimentation” with AI technology. Now, organizations are faced with serious budget growth, with CFOs “pushing for ROI-driven business cases,” he said.
Finance leaders should ask vendors for a proof of concept with “a very clear framework” for what the ROI will be, Etkin advised. Also, companies should also establish contracts with tiered pricing, so they aren’t paying the same rate per credit as they grow and need to use the AI tool more.
“My expectation is, over the course of the back half of 2026, we’re going to see a lot more resistance to these runaway credit consumption ordeals within these different functions, with obfuscated and ambiguous credit-based pricing,” he said.
About the author
Alex Zank
Alex Zank is a reporter with CFO Brew who covers risk management and regulatory compliance topics. Prior to CFO Brew, he covered the property/casualty insurance industry.
News built for finance pros
CFO Brew helps finance pros navigate their roles with insights into risk management, compliance, and strategy through our newsletter, virtual events, and digital guides.
By subscribing, you accept our Terms & Privacy Policy.