AI-native ERP

What is an AI-native ERP?

Definition

An AI-native ERP is a general ledger and finance system designed from the start around automation: transactions are categorised, reconciled and posted by models rather than by people keying entries, with the accountant reviewing. The term is used to separate new systems such as Rillet, Campfire and Light from incumbents that have added AI features to older architectures.

Also calledAI ERP

Compare the softwareBest general ledger software →All 14 ERP and general ledger vendors →

Named target
NetSuite
Category
ERP and general ledger
Vendors profiled
14
01

What makes an ERP AI-native

An AI-native ERP is one where the automation is part of the core ledger, not a feature added to it. A bolt-on AI feature sits on top of a system built for manual entry: it suggests a code, drafts a summary or answers a question, and a person still performs the posting. In an AI-native system the models do the posting and the person reviews exceptions.

Four criteria separate the two. First, the ledger is built around automated entry, so transactions arrive from banks, billing and payroll and are classified and posted without keying. Second, the data model is unified, with subledgers, entities and currencies in one structure rather than stitched together by add-ons. Third, reconciliation and close run continuously, so the books are current during the month and not assembled after it ends. Fourth, every automated entry is traceable: the system records what the model did and why, and the accountant can approve, correct or reverse it. A product that meets only the first criterion, such as an assistant layered over an older ledger, is a bolt-on.

02

How AI-native ERPs work

The new systems share a pattern: real-time books fed by direct integrations to banks, billing and payroll, revenue recognition and multi-entity consolidation built in, and agents that prepare the close. They target companies leaving QuickBooks or Xero and companies that would once have bought NetSuite. Funding has followed: Rillet reached a $1 billion valuation in 2026, Campfire raised $100 million in a year, DualEntry launched with $90 million and Lensing with $140 million. The incumbents answer with bundled AI: NetSuite's close manager and bank matching, Sage Copilot, Intuit Enterprise Suite.

03

Vendors to know

Rillet sells to high-growth companies, with a focus on SaaS revenue recognition and close. It raised a $70 million Series B in August 2025 led by Andreessen Horowitz and ICONIQ, then a $100 million Series C at a $1 billion valuation in August 2026, led by ICONIQ.

Campfire targets startups and growth companies that have outgrown entry-level accounting software. It raised a $35 million Series A led by Accel in June 2025 and a $65 million Series B co-led by Accel and Ribbit in October 2025, bringing total funding to about $100 million.

DualEntry aims at mid-market companies replacing legacy accounting systems. It emerged from stealth in October 2025 with a $90 million Series A led by Lightspeed Venture Partners and Khosla Ventures, with GV participating.

Light, based in Copenhagen, builds a general ledger for growth companies. It raised a $30 million Series A led by Balderton Capital in September 2025, bringing total funding to $43 million.

Quanta, based in San Francisco, sells a general ledger that keeps books continuously reconciled. It raised a $15 million Series A led by Accel in December 2025 and launched Prism, a reporting layer, alongside it.

Doss targets businesses that sell physical goods, with inventory, orders, procurement and fulfilment in one operations platform. It raised a $55 million Series B in March 2026 co-led by Madrona and Premji Invest.

04

Migration risk

Moving from a legacy ERP is a finance project before it is a technology project. The main risk sits in the data: opening balances, open receivables and payables, and the mapping of the old chart of accounts to the new one. Errors at cutover carry into comparative reporting and into the audit, so the closing trial balance of the old system should reconcile to the opening balances of the new one, and open subledger items should tie to their control accounts.

Finance teams also decide how much history to bring across. Many migrate opening balances and the two to three years of detail that auditors and lenders ask for, and archive the rest. Cleansing duplicate and inactive records is best done before the move, not during it.

Timing matters. Teams usually go live at a period end and close one month in both systems side by side, so the controller can compare results before the old system is retired. The parallel run adds work for a team that is also running the live close, so it is kept short. Integrations with banks, payroll and billing need rebuilding and testing, and approval workflows, user roles and controls need to be set up again and reviewed. Staff who have worked in the old system for years also need training on a model where they review entries instead of keying them.

05

Common questions

How is an AI-native ERP different from NetSuite with AI features?+

The difference is where the automation sits. AI-native systems are built so the ledger updates itself and people review; incumbents add assistants and agents on top of workflows designed for manual entry.

Who buys AI-native ERPs?+

Mostly technology and multi-entity companies from about 50 to 2,000 staff, often at the moment they outgrow QuickBooks or Xero or face a NetSuite renewal.

Which are the main AI-native ERP vendors?+

Rillet, Campfire, DualEntry, Light, Lensing, Doss and Quanta are profiled on Accountio; Digits and Puzzle sell autonomous ledgers to smaller companies.

06

AI-native ERP vendors

Part of Accountio’s accounting technology coverage · Glossary