Growth is usually a good problem to have, until the systems supporting the business can no longer keep up.
For finance teams, the warning signs often emerge gradually: another spreadsheet for reconciling accounts, another manual process for consolidating entity data, or another reporting layer outside the ERP. The software may still function, but employees must do increasingly more work around it.
Eventually, financial closes take longer, reporting falls behind decision-making needs, and leaders spend more time validating data than acting on it.
At that point, the question is no longer whether the ERP works, but whether it still works well enough to support how the business actually operates today.
To explore this inflection point, eWeek recently hosted a Trend Talk sponsored by Intuit Enterprise Suite, bringing together Corey Noles, AI expert and host of The Neuron, and Max Crampton-Thomas, managing editor and ERP expert. The discussion focused on what AI can realistically deliver inside financial workflows, where human oversight must remain in place, and how growing organizations should evaluate whether their ERP is still fit for purpose.
What emerged from the conversation was not a story about AI replacing ERP systems, but a more practical set of expectations for what “AI-native” should actually mean in finance. The key takeaway: adding AI features does not automatically make an ERP intelligent. Real value comes from whether AI can meaningfully reduce work while preserving visibility, governance, and accountability.
The following recap highlights the most important insights from the Trend Talk, and what they mean for finance and technology leaders deciding how to scale their systems for the next stage of growth.
Look at the work happening around the ERP
Organizations rarely experience a sudden ERP failure when they outgrow their system. More often, employees gradually build parallel processes around it.
A small finance team may initially manage exports, reconciliations, spreadsheet consolidations, and separate reporting layers. As the organization grows, those workarounds become harder to scale and increasingly dependent on the knowledge of specific employees.
“Spreadsheets aren't the problem. They become a problem when they're the glue holding the finance process together,” said Crampton-Thomas.
The concern, then, is not spreadsheet use itself. It is whether spreadsheets and other manual processes have become necessary to connect financial data, reconcile results, and produce a decision-ready view outside the ERP.
Finance and technology leaders should therefore look beyond whether the ERP remains operational. How much work happens outside it? Which processes rely on manual intervention? How long does it take to produce information the business trusts? These questions may reveal more about the system’s limitations than uptime or basic functionality.
Ask what the AI actually does
As AI becomes standard in enterprise software, asking whether a platform “has AI” provides little to no useful information. A better question is what the AI does within an actual financial workflow.
That work may include identifying inconsistencies, reviewing transactions, standardizing data across entities, detecting variances, supporting portions of the financial close, or explaining results using source data.
The distinction between an assistant and a more agentic system lies in whether the AI merely provides information or participates in the workflow.
“A standard assistant might tell you that these two accounts look inconsistent, but an agentic system could identify that inconsistency, recommend the correction, execute the approved workflow, and preserve a record of what actually happened,” Noles explained.
The difference is not simply greater automation. In finance, an active system must operate within established permissions and approvals while maintaining a traceable record of what changed, why it changed, and who authorized it.
They also noted that vendors should demonstrate recognizable financial workflows, not simply polished conversational interfaces. Buyers also need to understand which information the AI can access, what actions it can perform, and where human approval is required.
Enterprise AI should ultimately be judged by the work it completes, not by how convincingly its interface imitates human conversation.
Use AI to focus human attention
Finance professionals provide the most value when interpreting information and making decisions, not reviewing thousands of routine transactions to find a small number of unusual entries.
AI can help narrow that workload by processing structured data and directing attention toward exceptions that require closer review. The goal is to give employees a smaller, better-prioritized set of items on which to exercise judgment.
This creates a practical starting point for adoption. Organizations can identify repetitive processes in which AI could reduce routine review while escalating anomalies, material discrepancies, and uncertain decisions to qualified employees.
Human oversight remains essential. Finance teams must define the policies, thresholds, permissions, and approval structures under which the technology operates. AI may perform repeatable work within those boundaries, but accountability for consequential decisions remains with people.
Do not trade transparency for speed
A faster month-end close can provide leaders with a more current view of the business. However, speed has limited value if teams cannot verify how the system produced its results.
For every material action, users should be able to determine what source information the AI used, why it flagged or changed an item, what action it performed, and who reviewed or authorized it. The audit record should connect the outcome directly to the underlying data.
“In finance, explainability is not a nice-to-have for any material action. The team should see the source data and the audit record; otherwise you have traded a manual process for a completely different risk,” Crampton-Thomas explained.
Explainability cannot be added after automation is deployed. It is part of the control framework that makes automation usable and trustworthy.
The objective is stronger control with less manual effort, not faster processes with less visibility.
Move finance from assembly to analysis
Disconnected processes create particular challenges for multi-entity organizations. Financial data may be managed separately across entities, leaving finance teams to consolidate the results while accounting for intercompany activity and eliminations.
When that work relies on exports and spreadsheets, producing an organization-wide view takes time. A report can be accurate but still arrive too late to influence the decision it was intended to support.
A connected system can provide a shared view across entities. AI can assist by reviewing consolidated information, identifying meaningful changes, and linking findings to the underlying transactions for validation.
“It shifts finance’s time from assembly to analysis,” Crampton-Thomas said. “That’s when finance moves beyond reporting the business and starts helping to steer it.”
The larger opportunity is to give finance teams more time to determine what the numbers mean, why conditions are changing, and what the organization should do next.
Define implementation success before pursuing speed
Cloud deployment, migration tools, guided onboarding, and AI assistance may reduce some implementation work. They do not eliminate the need to align data, processes, integrations, controls, testing, training, and change management.
When vendors promise rapid implementation, buyers should clarify what “live” means. Does it refer to one entity, functioning core financial capabilities, or adoption across the entire organization?
AI may help map and standardize information during migration, but organizations still need clear data ownership and validation requirements. Otherwise, they risk transferring existing inconsistencies into a newer system.
A practical model for AI-native ERP
The Trend Talk presented a grounded standard for AI-native ERP: The technology should reduce repetitive work, direct employees toward what requires attention, and preserve the controls required for financial operations.
Intuit Enterprise Suite applies that model to the needs of growing, multi-entity businesses by helping unify financial data, automate key workflows, and improve visibility across the organization.
If your business is ready to reduce manual work, simplify multi-entity complexity, and gain a clearer view of financial performance, learn more about Intuit Enterprise Suite.