The Modern Finance Glossary

The language of finance is changing. As AI reshapes how accounting teams close the books, manage compliance, and deliver insights, new terms are entering the conversation. Some are industry-wide. Others are specific to the platforms and architectures driving this shift.

This glossary defines the most important terms in modern finance and accounting technology, from foundational concepts to the specific tools and features that Rillet uses to power AI-native financial operations.

Core Concepts

AI-Native ERP

An AI-native ERP is an enterprise resource planning platform that was designed and built with artificial intelligence as a foundational element of its architecture, rather than adding AI features on top of an existing legacy system. In an AI-native ERP, the data model, integrations, and workflows are structured so that AI can operate directly inside the system, reading and acting on financial data with full context.

This is distinct from "AI-enabled" or "AI-enhanced" ERPs, which typically bolt AI capabilities (such as chatbots or analytics dashboards) onto software that was originally designed without AI in mind. In those systems, AI operates in a separate layer and often lacks access to the structured metadata it needs to perform accounting tasks accurately.

Rillet is an AI-native ERP. Its general ledger, integrations, and automation workflows were built from the ground up for AI, which is why its Aura AI agents can draft journal entries, run reconciliations, generate accruals, and perform flux analysis directly inside the ledger.

Related terms: Aura AI, Continuous Close, Legacy ERP

Continuous Close

Continuous close is an accounting architecture in which transactions are processed, journal entries are posted, accounts are reconciled, and anomalies are flagged automatically and continuously throughout the accounting period, rather than being compressed into a batch process at month-end.

In a continuous close architecture, the month-end close becomes a confirmation step rather than a multi-day project. The books are substantially closed every day because the system handles the work as data arrives, not because the accounting team is working harder or faster.

This is fundamentally different from the traditional periodic close used by legacy ERPs, where amortization, reconciliation, accruals, and flux analysis all queue up and get processed sequentially during a compressed close window.

Rillet's continuous close architecture is the foundation of its platform. It enables features like real-time reporting, automated accruals, and zero-day close by ensuring that financial data is always current, reconciled, and audit-ready.

Related terms: Zero-Day Close, Batch Processing, Month-End Close

Zero-Day Close

A zero-day close refers to completing the financial close process (balancing the books, reconciling accounts, and preparing financial statements) on the same day a reporting period ends, rather than taking days or weeks after period-end.

Achieving a zero-day close requires a combination of continuous data processing, automated reconciliation, AI-assisted journal entries, and native integrations that eliminate manual data gathering. It is the practical outcome of a continuous close architecture.

Rillet is designed to help finance teams achieve a zero-day close by automating the majority of close tasks throughout the period, so that when the period ends, the books are already substantially closed and month-end becomes a review and confirmation step.

Related terms: Continuous Close, Close Management, Aura AI

Batch Processing (in Accounting)

Batch processing is the traditional method used by legacy ERP systems to handle financial data. In a batch processing architecture, transactions, reconciliations, journal entries, and other accounting tasks accumulate throughout the period and are then processed together in a compressed window, typically during the month-end close.

This approach creates several problems for finance teams: data is stale for most of the period, close week becomes a high-pressure sprint, and AI tools connected to the system can only work with data that was last synced during a batch run.

Batch processing is the architectural opposite of continuous close. The distinction matters because it determines whether AI features in an ERP can operate on current data or are limited by the same delays as the rest of the system.

Related terms: Continuous Close, Legacy ERP, Month-End Close, Periodic Close

Legacy ERP

A legacy ERP is an enterprise resource planning system that was designed and built before the current era of cloud computing, AI, and real-time data processing. Examples include Oracle NetSuite, SAP, and Sage Intacct.

Legacy ERPs typically share several characteristics: they rely on batch processing rather than continuous data flows, their integrations depend on third-party middleware, their data models were not designed for machine learning, and adding AI capabilities requires bolting on separate tools that operate outside the core ledger.

While legacy ERPs can be enhanced with AI features, the underlying architecture limits what AI can accomplish. The AI layer inherits the same constraints (batch processing, middleware data loss, rigid data models) as the rest of the system.

Related terms: AI-Native ERP, Batch Processing, AI-Enabled vs. AI-Native

Month-End Close

The month-end close is the process by which a business finalizes and reconciles all financial transactions for the previous month. It typically involves reviewing and posting journal entries, reconciling bank accounts, processing accruals and amortization, performing intercompany eliminations, running flux analysis, and preparing financial statements.

In traditional accounting, the month-end close is a sequential, labor-intensive process that can take anywhere from 5 to 15+ business days. Modern approaches, particularly those built on a continuous close architecture, aim to reduce or eliminate the close as a distinct event by processing these tasks automatically throughout the period.

Related terms: Continuous Close, Zero-Day Close, Close Management

Rillet Platform and Features

Aura AI

Aura AI is Rillet's built-in artificial intelligence system. Unlike generic AI chatbots that sit on top of an ERP, Aura AI consists of specialized agents, each trained on specific accounting workflows and connected directly to Rillet's general ledger.

Aura AI agents include:

• Flux Analysis Agent: Runs flux analysis at the transaction level, automatically tracing variances back to specific vendors, timing differences, and unusual transactions.
• Accruals Agent: Analyzes historical vendor data and billing patterns to predict and generate accruals, with plain-language rationale explaining why each accrual was created.
• Journal Entries Agent: Searches existing entries, explains their purpose, and creates new entries on command based on upstream agent findings.
• Reconciliation Engine: Uses proprietary machine learning models to automatically match 95%+ of incoming bank transactions with invoices and bills.

Aura AI applies accounting logic first, then surfaces results. Finance teams can interact with Aura in plain English to ask questions, request data summaries, generate reports, and instruct Aura to book journal entries directly.

Related terms: AI-Native ERP, Aura Flows, Continuous Close

Aura Flow

Aura Flow is Rillet's AI workflow builder. It allows finance teams to orchestrate multi-step financial processes where Rillet decomposes a complex task into individual steps, builds a plan, and executes the entire workflow end to end.

For example, a 13-week cash forecast using Aura Flows works like this: Rillet automatically pulls open invoices, open bills, current cash balance, and historical trends, then synthesizes a forecast with a graph, ready to share. Users can watch progress in real time and inspect intermediate results at any step.

Aura Flow represents a shift from "ask the AI a question" to "tell the AI to do the work." It moves Rillet's AI capabilities from insight delivery to task execution.

Related terms: Aura AI, AI Workflow, Continuous Close

Close Management

Close management in Rillet refers to the built-in system for organizing, automating, and tracking the month-end close process. It includes automated close checklists, task assignment and approval workflows, error detection, and real-time visibility into close progress.

Rillet's close management is designed to work in tandem with its continuous close architecture. Because most close tasks are handled automatically throughout the period, the close management system focuses on review, approval, and exception handling rather than manual data processing.

Related terms: Continuous Close, Zero-Day Close, Month-End Close

Native Integrations

Native integrations are data connections between Rillet and other business tools that are built and maintained in-house by Rillet's engineering team, rather than relying on third-party middleware or iPaaS connectors.

Rillet offers 25+ native integrations with tools including Stripe, Salesforce, HubSpot, Ramp, Brex, Rippling, BILL, and others, with 20+ more in the pipeline. These integrations are designed to preserve the structured, metadata-rich data that Rillet's AI needs to operate accurately.

This matters for AI performance because third-party connectors often strip out metadata during data transfer, leaving the AI with incomplete context. Rillet's native integrations ensure that when data flows into the ledger, it arrives with the full context (entities, currencies, departments, vendor details, contract terms) that Aura AI needs to do its job.

Related terms: AI-Native ERP, Headless ERP, Open API

Headless ERP

A headless ERP is an enterprise resource planning platform built with an API-first architecture, designed to be accessed and extended programmatically as well as through a traditional user interface. The "headless" design means the backend logic and data layer are decoupled from the frontend, allowing other systems, custom applications, and AI tools to interact with the ERP directly.

Rillet's headless design and open API allow finance teams to build custom integrations, automate workflows, and connect AI tools (like Claude via MCP) directly to their financial data without being limited to the features available in the UI.

Related terms: Open API, MCP (Model Context Protocol), Native Integrations

AI and Integration Concepts

Model Context Protocol (MCP)

The Model Context Protocol (MCP) is an open protocol, originally developed by Anthropic and now governed by the Linux Foundation, that standardizes how AI models connect to external tools and data sources. MCP creates a universal adapter between AI assistants (like Claude) and business systems (like ERPs, CRMs, and databases).

For finance teams, MCP means an AI assistant like Claude can connect directly to an ERP, pull real financial data, and help work with it in plain language, without exporting CSVs or copying data between tools.

Rillet has a published MCP server that connects to Claude, giving the AI direct access to Rillet's general ledger data. This enables workflows like asking Claude to pull a trial balance, analyze vendor spending trends, or draft a board-ready financial summary using live data from Rillet.

MCP is only as useful as the data it connects to. An AI-native ERP like Rillet, with structured data and native integrations, provides the context that AI models need to return accurate, actionable results. Legacy ERPs connected via MCP often expose raw tables without the structured metadata AI needs to reason about financial data.

Related terms: AI-Native ERP, Aura AI, Headless ERP, Open API

AI Agents (in Accounting)

AI agents are specialized, task-oriented AI systems designed to perform specific workflows autonomously or semi-autonomously, rather than simply answering questions. In accounting, AI agents handle defined tasks like reconciliation, accruals, flux analysis, or journal entry creation.

AI agents differ from AI chatbots in a critical way: chatbots respond to questions, while agents take action. An AI chatbot might tell you that R&D spend increased 15% month-over-month. An AI agent traces the variance to three specific vendor invoices, drafts the flux analysis narrative, and prepares a journal entry for your review.

Rillet's Aura AI is built on a multi-agent architecture. Each agent is trained on a specific accounting workflow and can hand off findings to other agents when needed (for example, the Flux Analysis Agent can pass its findings to the Journal Entries Agent to draft a corrective entry).

Related terms: Aura AI, Aura Flows, AI-Native ERP

AI-Enabled vs. AI-Native

AI-enabled (also called AI-enhanced or AI-added) refers to software that was originally built without AI and has since had AI features added to it. The AI operates in a separate layer, typically as a chatbot, analytics dashboard, or copilot that sits on top of the existing system.

AI-native refers to software that was designed and built with AI as a core architectural component from the start. The AI operates inside the system, with direct access to structured data, business logic, and workflows.

The distinction matters because AI-enabled systems inherit the limitations of their underlying architecture. If the ERP batches data, the AI works with stale data. If integrations lose metadata, the AI loses context. An AI-native system avoids these constraints because the data model was designed for AI from day one.

Related terms: AI-Native ERP, Legacy ERP, Aura AI

Financial Operations Terms

Flux Analysis

Flux analysis is the process of comparing financial data across periods to identify and explain significant variances. It is a standard part of the month-end close process and is typically required for audit readiness.

Traditional flux analysis is manual and time-consuming: accountants compare line items period over period, identify material variances, then investigate and document the reasons behind each one.

Rillet's Flux Analysis Agent automates this process at the transaction level, tracing variances to specific vendors, timing differences, and unusual transactions. It generates plain-language explanations that are ready for review, eliminating hours of manual investigation.

Related terms: Aura AI, Month-End Close, Close Management

Revenue Recognition (ASC 606)

Revenue recognition is the accounting principle that determines when and how revenue is recorded in financial statements. ASC 606 is the current U.S. GAAP standard for revenue recognition, requiring companies to recognize revenue based on the transfer of goods or services to customers, measured by the amount of consideration expected to be received.

For SaaS companies and businesses with complex pricing models (subscription, usage-based, milestone), ASC 606 compliance requires careful tracking of contract terms, performance obligations, and timing.

Rillet automates ASC 606 revenue recognition directly from contracts. Revenue treatment stays consistent, auditable, and up to date across pricing models including flat rate, usage-based, and milestone billing, without requiring manual spreadsheet calculations.

Related terms: GAAP Reporting, Contract-to-Cash, AI-Native ERP

Multi-Entity Consolidation

Multi-entity consolidation is the process of combining financial data from multiple legal entities (subsidiaries, international offices, or business units) into a single set of consolidated financial statements. This process involves currency conversions, intercompany eliminations, and reconciliation across entities.

For growing companies, multi-entity consolidation is one of the most complex and error-prone parts of the close process, particularly when operating across currencies and jurisdictions.

Rillet handles multi-entity consolidation natively, with automated currency re-evaluations, intercompany journals, and reporting that lets finance teams switch between consolidated and subsidiary views instantly.

Related terms: Intercompany Elimination, Month-End Close, GAAP Reporting

Intercompany Elimination

Intercompany elimination is the process of removing transactions between related entities (such as a parent company and its subsidiaries) during financial consolidation. This prevents double-counting of revenue, expenses, assets, and liabilities in the consolidated financial statements.

Rillet automates intercompany eliminations as part of its multi-entity consolidation process, ensuring that consolidated financial statements accurately reflect only external transactions.

Related terms: Multi-Entity Consolidation, Close Management, GAAP Reporting

Contract-to-Cash

Contract-to-cash refers to the end-to-end financial workflow from contract creation through invoicing, payment collection, and revenue recognition. It encompasses the full lifecycle of a customer financial relationship.

Rillet natively syncs with CRMs like Salesforce and HubSpot, automating the entire contract-to-cash process: contract creation, invoicing, payment tracking with automated reminders and secure payment links, and revenue recognition, all connected to live accounting data in the general ledger.

Related terms: Revenue Recognition (ASC 606), Accounts Receivable, Native Integrations

GAAP Reporting

GAAP (Generally Accepted Accounting Principles) reporting refers to financial statements and reports prepared in accordance with the standardized accounting rules used in the United States. GAAP compliance is required for publicly traded companies and is expected by investors, auditors, and lenders.

Rillet serves as the single source of truth for both GAAP and investor reporting, ensuring that all data remains consistent by design. Financial statements, management reports, and investor updates are all generated from one platform with no manual reconciliation between reporting sets.

Related terms: Revenue Recognition (ASC 606), Audit Readiness, SaaS Reporting

SaaS Metrics and Reporting

SaaS metrics are the key performance indicators specific to software-as-a-service businesses, including Annual Recurring Revenue (ARR), Monthly Recurring Revenue (MRR), churn rate, customer retention, and unit economics.

Rillet provides native SaaS reporting with drill-down dashboards, retention cohort tracking, budget vs. actual comparisons, and scenario modeling, all generated from the same ledger data used for GAAP reporting. This eliminates the common problem of having one set of numbers for accountants and another for the board.

Related terms: GAAP Reporting, Investor Reporting, Contract-to-Cash

‍