Solvimon says companies are adopting more complex pricing models as AI changes how software is billed. According to the billing platform, usage-based charges have risen sevenfold since 2025.

Its data points to a shift away from flat monthly fees and seat-based pricing towards combinations of subscriptions and consumption-based charges. The fastest-growing software and AI businesses now use nearly five pricing structures on average, up from three over the past year.

That shift is putting pressure on finance teams as AI products create value in ways that do not fit neatly into older billing systems. Businesses are trying to bill for credits, tokens, usage and outcome-based measures alongside standard subscriptions, increasing the risk of invoicing errors and missed revenue.

A Bain & Company analysis of more than 30 established SaaS vendors found that 65% have already introduced hybrid pricing, adding AI usage or outcome metrics to traditional seat-based models. Solvimon argues that the spread of these structures is making billing one of the more fragile parts of the revenue cycle for companies with expanding product lines and more tailored commercial contracts.

Research from MGI found that SaaS companies can lose between 1% and 5% of annual recurring revenue each year through billing leakage caused by errors that repeat in each billing cycle until detected. For chief financial officers, that is bringing greater scrutiny to systems once treated as back-office tools rather than strategic infrastructure.

Oracle customers have been warned to watch for changes in support and pricing as Larry Ellison’s company makes huge datacenter spending commitments to support its AI ambitions.

Advisors in Oracle negotiations said support levels are declining, while costs might increase as discounts disappear. Others have warned of a tougher approach to contract negotiations.

Oracle is aggressively expanding AI datacenter capacity to serve demand tied to customers and partners including OpenAI, xAI, Meta, Nvidia, and AMD. The deal for OpenAI alone is set to be worth about $300 billion over five years.

Widespread reports of layoffs affecting thousands of staff came as investor concerns grew over how Oracle would finance its AI datacenter build-out, while borrowing heavily to support the expansion.

In March, Oracle increased the estimated cost of its fiscal 2026 restructuring plan by $500 million, taking the total to as much as $2.1 billion, as observers anticipated broader job cuts.

Oracle previously raised $18 billion in debt and has since outlined plans to raise far more, as it funds datacenter investments aimed at meeting surging demand from AI model builders and enterprise customers. Ratings agency Moody’s warned of further borrowing and negative free cash flow as Oracle ramps up its AI infrastructure investment.

Nick Walter, CTO of professional consulting at commercial advisory firm House of Brick Technologies, said that there were plenty of anecdotal reports that Oracle support outcomes are getting worse, meaning slower service, less skilled help, and “more Oracle Support responses that were obviously written by AI.”

House of Brick had also seen a recent surge in customers asking for help in reducing Oracle Cloud Infrastructure (OCI) usage or help in negotiation with Oracle over long-term OCI deals, he said. “Many customers who adopted OCI in the last three to five years are suddenly finding out that their renewals are much more expensive because Oracle has eliminated or reduced enterprise discounts, and they are being forced close to public listed pay-as-you-go prices.”

Customers were also seeking help in response to pressure on Java, with audits and sales pushes and “scary-sounding threats.”

“This is not a new phenomenon but pressure continues and Java is the main front in Oracle’s war against their customer’s pocketbook,” he said.

Craig Guarente, founder and CEO of Palisade Compliance, said that Oracle was no longer simply putting pressure on customers to license Java; it was conducting official audits. “We’ve definitely seen in the last few months a return to classic Oracle auditing practices,” he said.

In 2023, Oracle changed its Java SE subscription model, shifting from a per-user or per-processor basis to per-employee. Critics called the move “predatory” as organisations that were using little Java but had a large number of employees could be hit hard by the cost increase. Later that year, research from Gartner showed that costs could be between two and five times greater under the new licensing model for using the same software.

Guarente also warned of price hikes at OCI contract renewal. “If you’re not adequately protected, then we’re getting 9 percent or 10 percent for long term lock in. It all goes back to that original purchase. You had some flexibility in there if you were thoughtful in how you negotiated those contracts. But if you signed the local standard agreement, there’s no price protection.”

Eric Guyer, founder of Oracle advisory and consulting firm Advizant, pointed out in a social media post that the margins in software support were much higher than in datacenters.

“This means the value to Oracle of your software support payments is increasing, desperately so, whereas the value to you is diminishing, and at an increasing rate. I mean this both technically via waning innovation and the more than 8 percent annual inflationary adjustments,” he said.

Customers are also concerned about the long-term impact of the job losses on their relationship with Oracle, Guyer said.

“The trend in license negotiations is only indirectly related to Oracle’s shift for now. The layoffs mean companies have lost one or more of their sales contacts. But for apps or the more strategic long-standing relationships, this means customers are wondering what this loss of institutional knowledge will mean for them. I’m not hearing customers wonder if Oracle is de-prioritising its legacy software business just yet, although I see this as an obvious outcome,” he said.

Does your AI agent count as an employee?

At a recent conference, Microsoft executive Rajesh Jha floated a provocative idea. In a future where companies deploy fleets of AI agents, those agents may need their own identities — logins, inboxes, and even seats inside software systems. If so, AI wouldn’t shrink software revenue. It could expand it.

“All of those embodied agents are seat opportunities,” Jha said, envisioning organisations with more agents than humans — each effectively a user that must pay for a software license, or “seat” in industry lingo.

That’s a radical twist in the SaaS pricing debate rattling companies like Microsoft, Salesforce, and Workday. Investors worry AI could hollow out seat-based pricing, the backbone of enterprise software. If one human can manage dozens of agents, why pay for dozens of licenses?

Jha’s answer: because those agents are the new users. A company with 20 employees might buy 20 Microsoft 365 licenses today. If each employee gets five AI agents, and the workforce shrinks to 10 people, that could still mean 50 paid seats.

Not everyone buys it.

Nenad Milicevic, a partner at AlixPartners, sees the opposite. AI agents will reduce the number of humans interacting with software, slashing licenses. Instead of 20 employees, you might have one person overseeing a handful of agents. That shift would pressure vendors and empower customers to push back on pricing that no longer makes sense.

Milicevic argues the winners will be open platforms. Companies could charge extra for machine-based access, but risk losing customers to software rivals that let agents operate freely.

Which brings us back to the core tension: if AI agents are just extensions of you, charging extra feels like double billing. If they’re autonomous workers, paying for them may be inevitable.

The answer could define the next decade of software economics.

Is Copilot a serious productivity tool, or just a party trick? Microsoft’s heavy promotion of its AI is facing criticism because its terms of use say Copilot is for “entertainment purposes only.”

Last fall, the company quietly updated the Copilot Terms of Use to note that “Copilot is for entertainment purposes only. It can make mistakes, and it may not work as intended. Don’t rely on Copilot for important advice. Use Copilot at your own risk.”

The agreement adds: “We do not make any warranty or representation of any kind about Copilot. For example, we can’t promise that any Copilot’s Responses won’t infringe someone else’s rights (like their copyrights, trademarks, or rights of privacy) or defame them. You are solely responsible if you choose to publish or share Copilot’s Responses publicly or with any other person.”

Software companies are leaving money on the table because their core financial systems haven’t kept pace with the way they sell pay-per-use services, which often now incorporate AI capabilities.

According to a survey of 350 execs who work for software vendors, conducted by PwC UK and billing infrastructure biz m3ter, 44 percent of UK business leaders say they have trouble measuring the consumption of usage-based software.

PwC and m3ter claim this measurement blind spot leads to revenue leakage – the gap between the value of what the company sells and the amount actually billed.

This can take the form of failure to capture usage data, like a subscription overage fee that isn’t tracked. Or it might reflect out-of-date pricing information, or billing calculation errors.

M3ter claims that between 4 and 7 percent of annual recurring revenue is potentially at risk from unsophisticated bookkeeping.

AI services can compound the problem, with the pricing of AI software and services often opaque and difficult to predict.

With regard to the way that AI complicates billing for companies that use it in their products, the research points to PwC’s 2026 CEO survey that found “only 30 percent of companies surveyed reported increased revenue from AI in the last 12 months.”

“The monetisation opportunity exists, but it is clear that converting AI capability into sustainable, auditable revenue remains a challenge,” the PwC/m3ter report states.

“A useful comparison is with the cloud infrastructure sector,” Griffin Parry, CEO and co-founder of m3ter, said in an email. “They had similar challenges and over time they developed greater transparency, via both better pricing design and better accessibility of usage and spend data.”

But the survey results aren’t solely reflective of black-box AI billing – sometimes it’s just inefficient business processes like tracking multiple product lines that combine seat-based and usage-based billing in a spreadsheet.

The PwC/m3ter survey found that 87 percent of respondents reported that their billing systems and their ERP or general ledger systems were not integrated. And 48 percent said there was no integration between billing and CRM systems.

Parry argues that transparency is a key theme for corporate customers, in terms of pricing design and the accessibility of usage and billing data.

“Good usage pricing should be easy for customers to understand, easy for them to predict (i.e. they can estimate how much they’d pay in real world scenarios), and with the variable element attached to a good proxy for value (i.e., so that if they use more of your service, they’ll associate that with being successful),” said Parry. “That’s one way of interpreting transparency, the opposite of opaque.”

Parry went on to say that customers won’t be satisfied with one monthly invoice that summarises usage-based spending.

“Instead, they’ll want to be able to access information about their usage and spend on a near-real-time basis, and for this data to be detailed – i.e., they want access to a usage and billing dashboard, as, say, AWS customers do,” he explained. “That’s the other way of interpreting transparency, access to the data that drives bills.”