Procurement technology is entering a new phase, as enterprises move from platforms that track activity to systems that can initiate action. For years, procurement software was built around records and workflows: first enterprise resource planning tools, then source-to-pay suites that improved visibility, standardised processes and gave procurement a more strategic role. Yet, as Ardent Partners’ chief research officer Andrew Bartolini and Zycus’ director of product manageme...
Continue Reading This Article
Enjoy this article as well as all of our content, including reports, news, tips and more.
By registering or signing into your SRM Today account, you agree to SRM Today's Terms of Use and consent to the processing of your personal information as described in our Privacy Policy.
That limitation is now being challenged by agentic AI. Unlike conventional AI tools that mainly recommend, summarise or flag exceptions, agentic systems are designed to act within defined guardrails. In procurement, that means moving beyond dashboards and task queues towards software that can recognise a trigger, make a decision within policy and carry out the next step itself. The effect is to shift the function from managing process to owning outcomes.
Intake is emerging as the most important starting point for that change. Once viewed as little more than a request form or ticketing layer, it is increasingly being recast as the control point for procurement execution. Modern intake tools can interpret the context of a request, including category, budget, supplier history and strategic priority, before deciding whether to route it for approval, launch a sourcing event or direct the user to a guided buying path. In more advanced setups, the system can even take action autonomously inside preset limits. That reduces off-contract buying, improves compliance and removes friction for business users and procurement teams alike.
Sourcing is also being reshaped. According to technology providers and consulting groups working in this area, AI can already draft RFPs, normalise supplier responses and test award scenarios before a human category manager steps in. The bigger step is the emergence of autonomous negotiation agents, which can interact with suppliers, conduct structured exchanges and look for savings opportunities without constant supervision. Depending on the category, these tools are being linked to savings of around 2.5% to nearly 10%, while also allowing value creation to become a continuous process rather than something that happens only during periodic sourcing cycles.
That evolution is altering how procurement is judged. Instead of counting purchase orders, sourcing events or other activity-based measures, leaders are starting to focus more on outcomes such as cost reduction, shorter cycle times and lower risk. PwC says that this shift requires not only new tools but also better data readiness and changes to operating models. Deloitte has similarly warned that scaling multi-agent systems depends on strong governance, human oversight, interoperability and a reliable enterprise data core. TechRadar has also highlighted a familiar warning: many agentic AI projects fail when data is fragmented, outdated or poorly verified.
Vendors are already positioning their suites around this shift. Oracle describes its cloud procurement offering as an integrated source-to-settle platform with AI-enabled sourcing, supplier management and analytics. Ivalua, meanwhile, promotes a connected source-to-pay model that unifies intake, sourcing, supplier management and procure-to-pay processes in a single environment. The common theme is clear: procurement is moving away from disconnected automation and towards systems that can see across the process and act on what they find.
For procurement leaders, the promise is obvious, but so are the demands. Agentic AI will not succeed simply because it is available. It depends on clean data, clear policies, robust controls and a willingness to redesign how work is done. Even so, the direction of travel is unmistakable. Procurement is no longer being asked only to process spend more efficiently. It is being asked to act on it.
Source: Noah Wire Services



