What does it actually mean for an ERP to be AI-native and why does it matter for your finance team?
In this episode of the AI to ROI podcast, host Ray Rike sits down with Santiago Nestares (Santi), Founder and CEO of DualEntry, to unpack the real differences between legacy ERP systems and a ground-up AI-native platform.
Santi shares the origin story behind DualEntry, born from a painful nine-month ERP implementation at his previous company that cost a team of 12 and hundreds of thousands of dollars—and explains why simply adding AI features on top of old database architecture misses the point entirely. AI, he argues, isn't a feature you plug in; it's a design philosophy that must be embedded at every layer of the product.
Ray and Santi dig into one of the thorniest challenges in enterprise finance: the tension between probabilistic AI models and the zero-error standard that accounting demands. Their answer? Deterministic guardrails—approval workflows, permissioning layers, and audit trails—that let AI work freely in draft mode while keeping humans accountable for every posted transaction.
You'll also hear about DualEntry's "Next Day Migration" approach, including how the company uses AI to map and migrate every transaction (not just trial balances) in hours rather than months, giving prospects a live sandbox with their own data before they ever sign a contract.
What You'll Learn
Why adding AI to a legacy ERP is like "running an on-prem system with a CD on the cloud", and what truly AI-native architecture looks like instead
The difference between deterministic and probabilistic systems, and why accounting can't afford to get it right only 99.9% of the time without the right guardrails
How DualEntry's Next Day Migration works: AI-assisted mapping, atomic transactions, and a live sandbox demo using the prospect's own data
The real ROI of AI-native ERP from eliminating manual categorization drudgery to enabling multi-dimensional segmentation that surfaces hidden pockets of value and risk
How Dto build audit-ready explainability without being able to explain the AI itself - by tracing every decision back to a human approval
Why early-career finance professionals are "living the luckiest time" in the professionand how to lean into AI rather than fear it
Episode Topics at a Glance
00:00 — Welcome & guest introduction
00:51 — Santi's origin story: a nine-month legacy ERP nightmare that sparked DualEntry
02:52 — AI-native vs. legacy ERP: what's the real difference?
04:48 — Deterministic vs. probabilistic systems explained
06:49 — How to identify a truly AI-first platform vs. an AI add-on
10:07 — Next Day Migration: using AI to accelerate ERP transitions
14:11 — Implementation team design: finance practitioners + forward-deployed engineers
17:18 — Measurable ROI: from real-time bank feeds to AI-driven business insights
22:52 — AI explainability, audit trails, and the permissioning layer
25:06 — Rapid fire: CFO ROI variables, who owns AI ROI, and advice for early-career finance professionals
Podden och tillhörande omslagsbild på den här sidan tillhör
Ray Rike. Innehållet i podden är skapat av Ray Rike och inte av,
eller tillsammans med, Poddtoppen.
