Is Your Revenue Recognition Process Ready for New AI-Enabled Offerings?

AI is no longer just a feature companies bolt onto their products — it's becoming the pricing model itself. Vendors across software, industrial equipment, and services are shifting from flat subscriptions to usage-based consumption, outcome-based fees, prepaid AI credit pools, and equipment-as-a-service contracts built on AI-driven monitoring and predictive maintenance.

For finance and controllership teams, that shift raises an uncomfortable question: is your revenue recognition process actually built to handle any of this?

The Recurring Revenue Playbook Is Being Rewritten

Five distinct AI-era pricing patterns are now showing up in real contracts, often in combination:

• Usage/consumption pricing — billed per token, API call, or compute unit. In this approach, the platform usage is treated as one cumulative performance obligation, with the sum of usage-based fees during the month making up the monthly revenue.
  
  
• Outcome-based pricing — billed only when a defined result occurs, for example, the Zendesk customer service platform counts an "automated resolution" only after an issue is closed without a human agent.
  
  
• AI credit/token pools —  prepaid, abstract consumption units applicable across diverse AI capabilities. Industry leaders like Adobe and Microsoft Copilot Studio enforce non-rollover monthly expiration rules on unused credits. From a revenue management perspective, these expiring balances represent classic breakage under ASC 606 / IFRS 15. In a modernized SAP architecture, these credit pools are managed via SAP Subscription Billing, where period resets automatically trigger breakage Revenue Accounting Items directly into SAP Revenue Recognition —systematically converting unconsumed capacity into recognized, high-margin revenue without manual spreadsheet intervention..
  
  
• Industrial equipment-as-a-service and predictive maintenance — customers pay for uptime or output, not the asset. For example, Rolls-Royce's TotalCare model for aircraft engines that are billed per flying hour is a direct ancestor of many manufacturing equipment programs that bill customers per output, uptime, and usage.
  
  
• Hybrid/bundled models — a fixed platform fee layered with variable AI overage, which is becoming the norm at companies like Twilio, Databricks, HubSpot, Zendesk, Snowflake, SAP and others.

This isn't a fringe trend. Roughly 85% of surveyed SaaS companies have already adopted some form of usage-based pricing, and Gartner projects that at least 40% of enterprise SaaS spending will shift to usage-, agent-, or outcome-based pricing by 2030. True outcome-based pricing is still just 0.6% of primary pricing models today, which means most of this shift is still ahead for many companies.

No New Rules From FASB or IASB — Which Exacerbates the Challenges

The natural next question is whether accounting standard-setters have issued anything specific to guide companies through this. As of mid-2026, they have not. FASB's current technical agenda contains no project addressing AI, usage-based, or outcome-based revenue (FASB Technical Agenda Overview), and the IASB's Post-Implementation Review of IFRS 15 concluded in September 2024 with an explicit decision to take no further action on variable consideration and usage-based royalties — precisely the area that AI pricing stresses hardest (IASB PIR Feedback Statement).

Both boards, in other words, have looked at this and concluded that ASC 606 and IFRS 15 already provide sufficient principles. That's a reasonable technical conclusion, but it means the entire compliance burden for AI-enabled offerings falls on your judgment, your contract terms, and your systems — with no updated rulebook to lean on.

"No new standard" should not be mistaken for "no new risk." If anything, it raises the bar on documentation, since auditors and regulators will expect a defensible, consistently applied methodology built entirely on existing guidance applied to genuinely new facts.

Where AI Pricing Breaks the Old RevRec Playbook

For most SaaS and subscription businesses, revenue recognition has been comfortably mechanical: recognize the fee ratably over the contract term. However, AI-era pricing routes revenue through several of the more judgment-intensive corners of ASC 606 and IFRS 15 instead.

• Variable consideration gets harder to estimate. New AI offerings often launch with little or no usage history; exactly the condition that triggers the variable consideration constraint under ASC 606-10-32-12, which limits how much unpredictable consideration a company can recognize before uncertainty resolves.

• "Series" determination becomes a high-stakes call. Whether usage or outcome fees qualify as a single "series" performance obligation determines whether a company can use the variable-consideration allocation exception. Small contract details, for instance whether the available usage resets monthly versus annually, can flip the conclusion entirely.

• Stand-ready versus specified-outcome obligations change the timing. Stand-ready versus specified-outcome obligations fundamentally dictate revenue timing. A contract dollar lands in entirely different fiscal quarters depending on whether an AI agent promises continuous time-based, ratable access or a confirmed fulfillment -driven performance result . In SAP Revenue Accounting, getting this distinction wrong isn't just an accounting classification error—it alters when recognized revenue reaches the General Ledger, directly impacting earnings quarterly predictability.

• AI credits bring gift-card types of accounting into enterprise software. Prepaid, expiring credit pools require the same kind of unused-balance analysis historically reserved for gift cards and loyalty points. Whether credits expire monthly or carry forward determines which approach applies; and most companies haven't yet addressed how their own AI features work.

• Contract modifications arrive faster than the annual cycle. The sheer velocity of AI product roadmaps—marked by continuous agent rollouts, tier upgrades, and mid-term promotional pricing—has shattered the traditional annual renewal cycle. Under ASC 606 and IFRS 15, these frequent mid-term amendments represent distinct contract modification events that alter transaction pricing and standalone selling price allocation ratios. Managing this high-frequency change volume manually creates immense operational friction. By connecting SAP Quote-to-Cash processes directly to SAP RAR processes, subscription change operations dynamically feed updated Revenue Accounting Items  into the revenue engine—executing prospective or cumulative catch-up adjustments systematically without slowing down fiscal close.

• Remaining performance obligation disclosures become the new investor signal. Snowflake, for example, discloses that only 48% of its remaining performance obligations are expected to convert to revenue in the next twelve months, "based on historical customer consumption patterns," and states plainly that deferred revenue is no longer a meaningful indicator of future revenue on its own. As the usage-based mix grows, disaggregation and remaining performance obligation (RPO) disclosures carry more analytical weight than they used to.

Keys to Getting Ahead of AI-Driven Revenue Recognition Risk

• Inventory every AI pricing construct in your portfolio. Before you can apply judgment consistently, you need a complete list of every usage meter, outcome definition, credit pool, and equipment-monitoring contract currently in market; including exactly how each one resets, expires, or escalates.

• Build a documented contract-attribute decision framework. Because the "series" and stand-ready/specified-outcome conclusions hinge on specific contract terms, finance needs a repeatable framework, not a one-off judgment call per deal, to assure that deal desk and revenue accounting apply consistently at the point of contract structuring, not after the fact.

• Instrument billing and contract systems to flag modification events. Mid-term tier changes, added AI agents, and promotional pricing expiration all need to be automatically surfaced as potential contract modifications rather than discovered during quarter-close.

• Treat breakage and SSP allocation as ongoing processes, not annual exercises. Credit-pool consumption patterns shift as AI features evolve, so breakage estimates and standalone selling price allocations need to be revisited on a cadence that matches how fast the underlying product changes.

• Strengthen disaggregation and RPO disclosure processes now. As usage-based and outcome-based revenue grows as a share of the business, the quality and granularity of these disclosures will draw more scrutiny from auditors, investors, and analysts — well ahead of when any new accounting standard might eventually catch up.

How SAP Solutions Help Close the Gap

This is exactly the territory the SAP revenue management stack was built to operationalize. SAP Revenue Accounting and Reporting (RAR) remains the proven foundation for ASC 606 and IFRS 15 compliance, using a flexible rules framework to determine performance obligations and revenue allocation automatically rather than relying on manual, deal-by-deal judgment.

Building on that foundation, SAP Universal Revenue Management (Universal RevRec), combining Contract-Based Revenue Recognition (CBRR), Event-Based Revenue Recognition (EBRR), and Universal Parallel Accounting (UPA) on SAP S/4HANA Cloud Public Edition, is purpose-built for exactly the kind of consumption-driven, event-heavy revenue streams that AI pricing generates, unifying contract data in one place so that terms like reset frequency and credit expiration are captured correctly and applied consistently.

On the billing side, SAP Quote-to-Cash (formerly BRIM) and SAP Subscription Billing handle the metering, rating, and invoicing of usage- and outcome-based charges, feeding clean, well-structured transaction data into RAR and Universal RevRec rather than leaving revenue accounting to reconstruct intent from raw usage logs.

SAP Business AI and Joule strengthen modern revenue management from strategy through execution. Joule intelligently surfaces usage anomalies, flags contract modification events at the point of sales entry, and enables finance teams to model the revenue recognition impact of new pricing structures before launch. Complementing this, pairing SAP Analytics Cloud  with SAP RAR delivers dynamic reporting capabilities, giving finance and audit teams the deep drill-down visibility required to satisfy increasingly stringent ASC 606 and IFRS 15 disaggregation and Remaining Performance Obligation disclosure requirements.

Bramasol's Value Proposition

None of this is theoretical for Bramasol clients. The point is that the platform capability to manage this complexity already exists — the real work is in configuring it correctly for each company's specific AI pricing constructs.

Bramasol has implemented more SAP RAR deployments than any other partner, and that depth carries over directly into helping clients extend into Universal RevRec, Quote-to-Cash, and Subscription Billing as their offerings evolve toward AI-driven pricing.

Our team has spent years translating exactly this kind of judgment-heavy accounting guidance — performance obligation identification, variable consideration, breakage, disclosure — into configured, auditable SAP processes across revenue recognition, lease accounting, and subscription-based Digital Solutions Economy programs.

If your company is rolling out usage-based, outcome-based, or AI-credit pricing, you need to be confident your revenue recognition process is complete and your SAP configuration is keeping pace.

That's the conversation to have now, before the first audit cycle finds the gap for you.