DualEntry Introduces a New Vision for ERP With Agents as a Service

DualEntry is advocating an enterprise software pricing model in which businesses pay for work completed by AI agents rather than the number of employees using an application. Its proposed “Agents as a Service” approach challenges the per-seat model used across much of enterprise software for two decades; the provided article excerpt gives no pricing figures or adoption results.

Analysis

The investable question is not whether agent-based ERP can automate tasks; it is whether vendors can capture the value without undermining seat-based recurring revenue. If agents materially reduce the number of human users, seat-revenue exposure could be greatest for workflow and finance applications with limited pricing power. Conversely, incumbents such as SAP, Oracle, and Workday may defend economics by charging for transactions, automation, or outcomes—and have an advantage if customers prioritize integration with existing systems of record. A shift to outcome pricing could also make reported revenue less predictable and expose vendors to inference costs, implementation burden, and disputes over how “work completed” is measured.

The article does not establish customer adoption, realized productivity, or unit economics. Treat DualEntry’s positioning as a thesis to verify, not evidence of disruption. Near term, this is more likely a narrative catalyst than a standalone earnings catalyst. Over 1–3 months, watch incumbent pricing disclosures, renewal commentary, and customer evidence of production deployments. Over 6–18 months, the structural risk is greatest if buyers reduce paid seats before vendors successfully monetize agent usage. The contrarian point: finance workflows involve approvals, controls, and auditability, so humans may remain licensed users even as agents handle more tasks.

Key Decisions for Investors

• No direct trade on this announcement alone: the article provides no adoption, pricing, retention, or cost data, and DualEntry is not identified as a publicly traded company in the supplied data.
• Set an alert on enterprise-software earnings for evidence of seat contraction, changes to per-user pricing, or explicit agent/transaction revenue; distinguish announced pilots from paid production deployments.
• If seat reductions become visible without offsetting usage revenue, consider reducing exposure to seat-sensitive application software relative to diversified incumbents with systems-of-record distribution. Reassess if renewals and net retention remain resilient while AI monetization grows.
• Falsify the disruption thesis if customers retain seat counts and vendors demonstrate paid agent adoption with stable or improving software economics. Escalate it if multiple vendors report lower user counts, weaker renewals, or agent usage that adds cost without measurable revenue.