Let me start with a defense of the thing everyone loves to criticize. Your ERP — whether it’s NetSuite, SAP, Microsoft Dynamics, or a decades-old system your company has run for years — is very good at what it was designed to do. It records the transaction. It keeps the general ledger. It closes the period and it survives the audit. Companies run on it for a reason.
So this isn’t an argument against your ERP. It’s an argument about the work that lives around it — the work no ledger was ever designed to own.
The ERP records. It was never asked to decide.
Think about what actually happens when someone in your company buys something. A request is made. A contract is found — or should be. A PO is cut. An invoice arrives, and someone checks it against the PO, and against the receipt, and, if anyone remembers, against the contract that set the price in the first place. An accrual gets booked. A journal entry gets drafted. A reconciliation ties it all back to the ledger. A variance gets explained.
Your ERP faithfully records each of those events once they’ve happened. What it doesn’t do — what it was never designed to do — is decide them. It doesn’t read the contract and notice the invoice is 20% over the rate in §4.2. It doesn’t draft the accrual from the PO before the close. It doesn’t chase the missing receipt. Those decisions fall to people, working across screens and spreadsheets, stitching together systems that each hold one piece of the truth.
The ledger is a system of record. The work around it is a system of decisions. No ERP was ever built to be both.
That’s why the bolt-ons exist — and why they don’t solve it.
The market’s answer, for twenty years, has been to bolt specialized tools onto the ERP. Coupa for procurement. Bill.com or Tipalti for AP. BlackLine for the close. Each one is good at its slice. And each one is a side-car — a separate application, with its own data model, that sits alongside your ERP (NetSuite, SAP, Dynamics, whichever you run) and has to be constantly synced back to it.
The result is a stack that looks integrated and behaves fragmented. Four data models. Zero shared record. The procurement tool stores the contract; the AP tool never sees it. The AP tool processes the invoice; the close tool reconciles it a month later. Every hand-off between them is a place where a person re-keys data, a variance slips through, or an audit trail breaks. Even the vendors admit this about each other — the hand-off between procurement and payables is a well-known blind spot, and the “integration” is usually a one-way push that leaves the teams in silos.
So you end up paying the disconnect tax twice: once for the tools, and again for the people who reconcile what the tools can’t.
Your ERP deserves better company.
Here is the reframe. The problem was never that your ERP is bad. The problem is that everything asked to help your ERP has been another disconnected system. Your ERP deserves a partner that does the opposite — one that treats procurement, AP, contracts, and the close as a single connected process on one data model, and does the decision-heavy work end to end, then posts the clean result back to the system of record you already trust.
That’s what Vakari is. Not a replacement for your ERP — a connected finance layer that runs alongside it:
- The contract becomes structured data the moment it arrives, so its terms can price the request and validate the invoice and draft the accrual — the same record, working three times.
- An agent runs the intake, the three-way match, the exception triage, and the JE drafting end to end — deciding with accuracy, completeness, and policy compliance, not just automating a click.
- Every action carries provenance back to its source clause and writes an immutable audit event. Segregation of duties is structural. Audit-grade is a byproduct, not a feature.
- And your ERP stays the system of record. Vakari posts into it. It makes the ledger better; it never tries to be the ledger.
Prevention beats reconciliation. When the record is connected the whole way through, there’s nothing left to reconcile — because nothing was ever apart.
The proof your ERP was waiting for.
Boards are asking CFOs about their AI strategy, and the honest answer isn’t “that’s the ERP vendor’s job.” The Office of the CFO is where AI’s value and its discipline both get tested first — where a wrong number has consequences and an unexplained action fails an audit. That’s exactly why finance is the right place to prove real agentic AI, and exactly why it has to be done with the control architecture built in from the start.
Your ERP did its part — it captured the transaction and held the line on the ledger for years. It earned a partner that finally does the rest. That’s the work Vakari was built for, and it’s the reason we say it plainly: your ERP doesn’t need replacing. It deserves Vakari.
— Amir Jafari, Founder & CEO, Vakari.ai
This is one of a monthly series on finance operations and agentic AI. Whatever ERP you run — NetSuite, SAP, or Microsoft Dynamics — if this describes your stack, the product and thesis pages go deeper. Running Microsoft Business Central or Dynamics specifically? The Business Central and Dynamics pages are tailored to those environments.