A reading of NetSuite's 2026–2030 direction for the leaders who have to run the platform through it — written from the operations side, not the roadmap deck.
The views below are my own, drawn from Oracle and NetSuite's public roadmap communications and from my work running ERP estates. Oracle would frame parts of this more favorably; I welcome that conversation.
The shift NetSuite is actually selling
For the better part of two decades, NetSuite's pitch was straightforward: one cloud suite that records what the business does. The roadmap NetSuite now markets — its "Next" generation — reframes that ambition. The product is no longer positioned as a system that captures transactions for people to review; it is positioned as a platform that reasons over them and acts.
That reframing matters less as a slogan than as a budgeting signal. When a vendor stops investing in better data entry and starts investing in autonomous execution, the cost a customer carries quietly migrates — out of license line items and into the unmeasured manual work that the new capability is meant to absorb. For a CIO or CFO running a mid-market estate, the useful question is not whether the vision is real. It is which parts land on your operations team in the next 18 months, and which can wait. Three threads carry most of the weight: the infrastructure move underneath the suite, the agent layer on top of it, and a new interface that becomes mandatory faster than most teams expect.
The OCI migration is not optional, and the friction is real
NetSuite is moving its entire customer base onto Oracle Cloud Infrastructure and the Autonomous Database. This is presented as a "push-of-a-button" upgrade rather than a reimplementation, and for the data layer that is largely true — the SuiteCloud architecture is designed to preserve existing customizations and scripts through the move.
But "push of a button" understates what an IT team owns during the transition. Mandatory single sign-on configuration, IP allowlist updates, and connectivity validation across every integration do not configure themselves. The migration removes future tuning work; it adds near-term transition work. The teams that absorb it well are the ones treating the cutover as a discrete, scoped engagement with a defined checklist — not a background event that happens to them.
The agent layer: where the unmeasured cost lives
The clearer story in NetSuite's roadmap is the move from generative AI — last cycle's summaries and drafted text — to agentic AI: software that reasons, plans, and executes multi-step workflows with limited human oversight. NetSuite describes a fleet of role-specific agents:
- A close manager intended to monitor transactions through the month, post accruals and journal entries as it goes, and pull finance toward a "continuous close" — a four-day month-end or better.
- An exception agent that flags anomalies (duplicate vendor bills, miscoded journals) in real time rather than leaving controllers to scan thousands of lines after the fact.
- A supply-chain agent that reads demand signals and vendor history to surface disruption before a stockout forces a scramble.
The interface shifts to match. NetSuite's "Ask Oracle" assistant is positioned to replace static dashboards with intent-based, conversational interaction — ask for a customer renewal email and the system assembles contract dates, pricing, and support history into a draft.
Redwood becomes mandatory faster than teams plan for
The Redwood design system is the visible face of all this — a move from the suite's traditional lists-and-forms to consumer-grade workspaces (a unified SaaS 360 or Customer 360 timeline rather than a stack of saved searches), with progressive disclosure to reduce cognitive load for a newer workforce.
The line item leaders should circle is the timeline, not the aesthetics. Redwood is the default for new implementations now, and the classic interface is slated for deprecation as early as 2027, partly because newer AI features assume the new shell. That converts a UI refresh into a training and adoption deadline. The risk is not that the interface is worse; it is that an estate drifts toward a forced cutover with no internal enablement plan, and discovers the cost as a productivity dip across every desk that touches the system.
The commercial play: "free" AI as an economic moat
Oracle's most aggressive move is bundling these AI capabilities into the core subscription at no additional charge — a deliberate contrast with premium per-seat pricing on Microsoft's Copilot and SAP's Joule. The strategy is transparent and worth naming plainly: included AI lowers the apparent total cost of the suite, discourages buyers from bolting on third-party tooling, and increases data gravity. The more company data consolidates inside NetSuite, the sharper the agents get and the higher the switching cost climbs. That is a coherent vendor strategy. It is also a reminder that "no additional cost" describes the license, not the work of realizing the value — and that the renewal conversation, where NetSuite has drawn criticism for steep step-ups, is where the real price is set.
A related expansion sits alongside it: SuiteBanking, NetSuite's partnership-backed move (with HSBC) from recording payments to executing them — AP/AR automation that approves and runs ACH or virtual-card payments in-screen, with working-capital and factoring features signposted for later in the roadmap. For finance leaders it collapses tool sprawl; it also deepens the same dependency the commercial strategy is designed to create.
What a CIO or CFO should actually do
Stripped of the roadmap gloss, the decisions resolve cleanly:
- Scope the OCI move as a project, not an event. Assign the SSO, allowlist, and integration-validation work an owner and a checklist before the cutover window, not during it.
- Treat included AI as unrealized cost, not free upside. Audit where your finance and supply-chain teams still do repeatable manual work today, then decide — process by process — which agents you will actually operationalize. The benefit is in eliminated workload, not in the license you already hold.
- Put a Redwood adoption plan on the calendar now. With classic deprecation signposted around 2027, internal enablement is the difference between a smooth transition and a productivity hit.
- Go into renewal clear-eyed. "Included" AI is a reason to consolidate, not a reason to stop negotiating; model the multi-year contract value against the manual work you can genuinely retire.
The through-line is the same one running through every modern ERP roadmap: the platform wants to move from a record of what happened to an agent that acts. That is a credible direction. Whether it lowers your total cost of ownership or simply relocates it depends on how deliberately you manage the operational layer underneath — the repeatable, support-heavy work the roadmap assumes will quietly disappear, but only does when someone makes it.
Analysis by Ravi Madhavan, offered as one practitioner's reading of NetSuite's published 2026–2030 direction. Originally published on LinkedIn. Alternative reads welcome. Allari runs the ERP support layer for JDE, SAP, Oracle Fusion, and NetSuite estates — measuring the manual, repeatable work these roadmaps assume away and finding where it can be automated.
