RightRev Launches AI Agents to Tackle Revenue Accounting's Messiest Work

RightRev says it processes more than $44 billion in annual customer revenue, and on November 12 the company staked a claim to automate the work that surrounds those numbers. It launched RightAgents, a suite of seven AI agents that check contracts, explain balance changes, and prepare revenue accounting entries for human review.

The target is a real pain point. Finance teams increasingly juggle usage-based pricing, prepaid credits, and contracts that change mid-month — even when billing runs smoothly, each amendment can shift what the accounting requires. A customer might buy credits upfront, consume them over months, add a product mid-contract, or drop into a different pricing tier. Each event changes the accounting work; none of them stops the invoice from going out.

How does the system divide labor between AI and accountants?

RightRev's approach pairs AI assistance with a rules-based revenue engine. Agents gather information, investigate differences, and prepare work for review. Finance teams keep approval authority over everything that reaches the general ledger.

That division matters, and RightRev seems to know it. An AI assistant can produce a plausible explanation of a revenue movement without establishing that the underlying accounting is correct. The company's answer: give agents access to the records and calculations behind each number while tying accounting execution to approved policies.

The company's technical distinction is the separation between an agent's interpretation and the engine that calculates revenue. Its documentation describes deterministic execution — approved rules produce repeatable calculations, and agent outputs link back to supporting records. The caveat: determinism does not guarantee correctness. A consistently applied rule can still embody a mistaken policy, and a missing source event can still leave a gap. Review and scenario testing remain substantive work, particularly before a new contract treatment touches the ledger.

What is RightContext?

RightRev describes RightContext as the platform's data and context layer. It pulls together billing, ERP, and usage records, normalizes the inputs, and lets finance teams define business-specific context. The revenue engine then applies the customer's policies on top of that foundation.

The same context layer supports RightAgents and external AI tools connected through Model Context Protocol (MCP). The company says its agents can be accessed from tools including Claude and ChatGPT, operating within each user's existing permissions.

The practical value is continuity between a question and its evidence. If a controller asks why a balance changed, a useful answer must connect the movement to the relevant agreement, source transaction, and accounting rule. An assistant wired only to a final spreadsheet total would leave most of that chain missing.

What do the seven agents actually do?

The RightRev AI product page groups the agents around detecting problems, explaining results, and preparing changes:

• Anomalies identifies unusual variances, missing prices, and unexpected allocations for review.
• Contract Checklist extracts terms and obligations and checks them against an accounting checklist.
• Revenue Assistant answers plain-language questions using the organization's revenue records.
• Architect converts written policies into proposed rules that finance approves before activation.
• Use Case Modeling builds and runs tests for proposed changes.
• Flux connects period-to-period movements to contracts and drafts explanatory commentary.
• Reconciliation compares source systems, RightRev, and the general ledger, and explains differences.

The broader platform underneath supports bookings, usage, delivery events, milestones, and other configured triggers. It assigns accounting rules to individual performance obligations, allocates amounts across bundled contracts, and recalculates schedules when agreements are amended. Its stated scope covers ASC 606 and IFRS 15 revenue recognition, plus recurring and one-time arrangements, revenue waterfalls, journal entries, and standalone selling price analysis.

The seven agents shift preparation and investigation across the accounting period. Instead of starting every reconciliation at month-end, teams can review exceptions as contracts and usage records arrive, turning the close into a sign-off on accumulated work rather than a scramble.

In plain terms: the launch moves RightRev's automation from calculating revenue to the investigation and explanation that surround it, while leaving customers' existing billing and ERP systems in place. For finance leaders, the open question is whether that combination actually reduces reconciliation effort while preserving a clear path from each reported number to the evidence behind it — a claim the $44 billion figure alone cannot settle.