AI Fundamentally Changes Software Pricing. Most Financial Systems Aren’t Prepared
Jagan Reddy is the founder and CEO of RightRev.
GettyArtificial intelligence (AI) has changed SaaS economics, replacing predictable pricing with token- and outcome-based models that align price with delivered value.
Vendors and buyers stopped debating whether AI is necessary long ago. They’re racing to monetize it, but their financial systems aren’t built to produce on-time, audit-ready reports that account for the highly variable, nuanced elements of AI deals. Revenue you can’t count cleanly isn’t revenue yet.
SaaS accounting teams face revenue recognition judgments that only get harder as deals sign at scale. The root issue is structural: traditional financial systems were built for flat, predictable pricing, not usage, tokens and outcomes.
When pricing moves faster than the accounting behind it, accuracy becomes guesswork, and every close gets harder as audit risk rises. Teams have to rebuild accounting to close with real accuracy and confidence, not a best guess.
Teams face compounding pressure with each day spent struggling to reflect highly dynamic deals into traditionally rigid systems. Failing to fix this broken system creates considerable uncertainty:
• Customers buy a block of AI usage up front. Some burn through it and buy more partway through the contract. Others barely touch it. No two accounts behave the same way, which makes revenue almost impossible to predict.
• Money can take weeks to show up when fees are tied to outcomes. And the amount keeps moving as customers change how much they use the product and what they use it for.
• The vendor’s top line now rides on what each customer actually does. That raises the stakes on how every deal is structured, and it turns getting customers to value fast into a revenue lever, not just a success metric.
Organizations can’t afford to slow down, but finance teams need the right infrastructure to match the speed that deals are already moving at.
SaaS pricing will continue to adapt as vendors strive to capitalize quickly on the value these pricing models represent, the flexibility they provide to the sales team, and the advantage they yield in the marketplace.
• What are you actually promising? Access to an AI agent for a set period, or a specific result that the agent delivers? Those are two very different promises, and they get counted very differently once the deal is signed.
• How predictable is the money? When you charge for outcomes, you can’t know up front what you’ll earn. You estimate, then adjust as real usage comes in. The more of your pricing that floats, the harder your numbers are to pin down and the more room there is to get them wrong.
• What happens to what customers don’t use? Prepaid credits often go partly unused. If you haven’t decided the rule in the deal, your revenue ends up hanging on a question nobody answered.
• Can your systems keep up? This is what separates the field. Bundled AI deals pack several promises into one contract, such as a software subscription, a block of AI usage and a fee tied to results, each counted on its own timeline. Get this right, and you launch new pricing without hesitation. Get it wrong, and you slow down every time you try something new. In this market, slowing down is falling behind.
Firms including Deloitte and EY have shared process and operational considerations that can influence how you build new policies. Use this guidance, along with the four questions above and other relevant developments, to implement and adapt processes that reflect your organization’s unique needs.
Sales teams can rapidly pivot with new pricing to sign deals. Finance teams need to keep pace to fully support successful sales outcomes.
Scrutinize your current systems and processes, assuming the status quo will most certainly not be enough.
Audit your financial infrastructure. Where do finance and revenue workflows operate? Spreadsheets and disconnected tools create manual effort and potential errors, preventing the speed and scale you need. Move that work into systems that apply the same revenue rules to every deal, so nothing depends on who built the spreadsheet.
Assess usage as it happens. You can’t charge for what you can’t see. When pricing depends on usage and results, you need a live read on both or your billing and revenue is always a guess after the fact.
Structure audit-ready deals: Deal dynamics have often existed in scattered documents. When you make a judgment call on a contract, document why. A deal you can explain a year later is a deal that survives scrutiny.
Align pricing and revenue accounting teams. Pricing will remain dynamic as products evolve and customers ask for terms you haven’t offered before. Create a standing connection between your pricing team and revenue accounting team so finance can support a new pricing model the same day it is requested.
When your systems can handle any pricing model, you can say yes to any way your customer wants to buy.
The era of AI pricing is here. Any company selling software must reassess its revenue architecture to preserve the predictability and accountability necessary to run a successful business.
AI pricing is exposing the limits of legacy revenue systems faster than most teams expected. Get ahead of it now, and every new monetization model becomes an opportunity. Wait, and each one becomes a fire drill.
Forbes Technology Council is an invitation-only community for world-class CIOs, CTOs and technology executives. Do I qualify?
