A finance director at a 320-person manufacturing company pulls up this year's software budget next to last year's. The per-seat prices on the spreadsheet haven't moved much. The total at the bottom has climbed anyway, and nobody flagged it before the invoice landed.

That gap isn't a rounding error. It's what happens when software billing stops running on seats alone and starts running on how much a tool actually gets used, especially the AI features attached to it. Most of what gets written about this treats it as a reason to buy another platform to watch the numbers. That's not what actually closes the gap for a company this size. What closes it is someone reading the contract, checking it against what changed, and catching the shift before the renewal, not a new dashboard nobody has time to monitor.

What changed in how software gets billed

For most of the last decade, a SaaS budget was a multiplication problem. Seats times price per seat, maybe an annual increase, done. A finance team could build next year's number from this year's without much guesswork.

That math still works for plenty of tools. It stops working the moment a vendor adds a metered layer on top: AI credits, per-action pricing, usage-based add-ons that scale with how hard a team actually leans on the product. The seat price on the invoice can stay flat while the total moves, because the part that's actually changing isn't the price. It's the usage.

This isn't the same problem as an AI tool's own token bill growing month to month, which is a governance question inside a company's own AI stack. This is different: it's the everyday SaaS platforms already under contract, Salesforce, Notion, Microsoft 365, quietly restructuring what "the price" even means. A company can do everything right on AI governance internally and still get surprised by a vendor changing the rules on a tool it's used for years.

Why software spend is outpacing what budgets expect

Global software spending is on track to hit $1.468 trillion in 2026, up 15.5% from 2025's $1,271 billion, according to Gartner's July 2026 IT spending forecast. That's a faster growth rate than the year before, and among the categories Gartner tracks, only data center systems and infrastructure-as-a-service are growing faster than software itself.

None of that growth is evenly distributed across seats. A company isn't hiring 15% more people this year. What's growing is what each seat, and each AI feature layered on top of it, actually costs to run. A budget built by taking last year's software line and adding a flat percentage doesn't account for that, because the increase isn't flat. It shows up unevenly, tool by tool, tied to which vendors added metering and which didn't. Some vendors held the line. Others didn't.

Three places this breaks without a dedicated team

Forecasting. A budget built on last year's number assumes this year looks like last year. Once a vendor adds usage-based pricing to even one or two major tools, that assumption stops holding, and nobody notices until the actual bill arrives.

Renewal review. A pricing or packaging change buried in a vendor email rarely gets caught before it takes effect. It's not that finance or IT isn't paying attention. It's that nobody's job is specifically to read every vendor's fine print against what the company is actually using.

Ownership. Tracking vendor-by-vendor pricing changes across a stack of a few hundred tools isn't a task that fits neatly inside finance, IT, or procurement alone. So it tends to sit with nobody, which is a different problem than sitting with the wrong person. Nobody owns it.

Put together, these three gaps are why a company can do nothing wrong and still end a quarter with a software bill that doesn't match what it planned for.

It shows up a little differently for companies in India, the UAE, and Saudi Arabia. Most major SaaS and AI vendors bill in US dollars, so a pricing change lands on top of whatever currency movement already happened that quarter. A finance team juggling both isn't just tracking a vendor's new pricing tier. It's tracking two moving numbers at once, and most budgeting processes were built to handle neither.

Get your software spend benchmarked before it surprises you

CostRoom reviews your SaaS and cloud contracts against current market pricing, not last year's number.

Book a Demo Call

What buying more software doesn't fix

It's tempting to read all this as a reason to add another tool, something that promises a dashboard of every vendor's pricing in one place. But the data on software buying itself suggests that's not where most companies are struggling.

Seventy-seven percent of organizations plan to increase software spend in 2026, according to G2's 2026 Software Buying Trends Survey of over 3,300 decision-makers. And yet only one in three buyers manage to adopt new software without some kind of disruption or regret. Sixty-one percent hit implementation problems in the past eighteen months alone.

Spending more doesn't automatically buy clarity. A new platform is one more contract to track, one more vendor whose pricing can shift underneath it, one more login somebody has to remember to check. The gap between a budget and a contract doesn't close because there's more software watching the software. It closes when someone with the market context actually reads what changed and does something about it before the renewal, not after.

What a spend optimization layer checks that a platform alone doesn't

This is the distinction worth being direct about. A SaaS management platform gives a company visibility: a screen showing what's spent, where. That's useful, and it's also where most tools in this category stop. Visibility tells a finance team what happened. It doesn't renegotiate a contract, and it doesn't tell a 300-person company whether the new price it's being quoted is actually fair.

An optimization layer works differently. It means someone is actually checking a vendor's new pricing against what similar companies are paying, flagging a packaging change before the renewal date locks it in, and sitting in the negotiation when a number needs to move. It's not a dashboard feature. It's expertise applied to a specific contract, on a specific vendor, at a specific moment, which is exactly the part a platform can't do on its own.

For a company without a dedicated FinOps or software asset management function, that distinction matters more than it sounds. The choice usually isn't between a tool and doing nothing. It's between another system to check and someone whose job is already to check it.

Getting started

None of this requires knowing every vendor's pricing model by heart. It starts smaller: pulling the current contracts for the handful of tools with the biggest spend, checking each one for a metered or credit-based component that wasn't there at signing, and flagging renewal dates that fall in the next two quarters.

That short list is usually enough to see where the budget and the contracts have already drifted apart, and where they're about to. Nothing more elaborate is needed to start.

Start with an AI and SaaS spend analysis

CostRoom maps your current vendor pricing against the market before your next renewal.

Book a Demo Call

Frequently asked questions

Why is my software bill higher even though the price per seat hasn't changed? Many SaaS vendors now charge for usage on top of a seat price: AI credits, per-action fees, or metered add-ons. The seat price can stay the same while total spend rises with how much a team actually uses the tool.

What is hybrid or consumption-based SaaS pricing? It's a billing model that combines a fixed subscription fee with a variable charge based on usage, such as AI credits consumed or actions taken. It replaces or supplements the flat per-seat model most budgets are built to expect.

How do I forecast software costs when vendors charge by usage? Start by identifying which of your current vendors have added a metered component, then check actual usage trends against what's included in the contract before assuming next year looks like this year.

Do I need a dedicated FinOps team to manage this? Not necessarily. What's needed is someone tracking vendor pricing changes against actual usage and market rates, which can be a role inside the company or handled by an outside spend optimization layer.

Does a SaaS management platform solve this problem? It helps with visibility into what's being spent, but it doesn't negotiate a contract or catch a packaging change before a renewal locks it in. That part still requires someone actively checking and acting on it.

Related reading: