Procurement’s push towards artificial intelligence is running ahead of the operational changes needed to make it work at scale. Research by Ardent Partners, sponsored by Ivalua, suggests that many teams want to introduce agentic AI across Source-to-Pay while still relying on manual work and disconnected systems.
More than four in five procurement teams say tactical activities take up at least half of their daily capacity, according to the study. At the same time, more than fo...
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.
ur in five organisations intend to deploy agentic AI within three years in areas including sourcing, supplier risk, contract management, procurement intake and accounts payable.
That contrast points to a wider challenge than simply choosing new software. If AI is added to processes that remain fragmented, its ability to support decisions across the procurement lifecycle will be limited. The research draws on responses from 311 chief procurement officers and senior procurement leaders across 25 industries worldwide.
The distinction is between automating tasks and coordinating work across connected processes. Automation can make an individual activity faster or remove it from an employee’s workload. Orchestration links information and decisions across the sequence of procurement activity, so that sourcing, supplier risk, contractual obligations, purchasing and payment can inform one another.
In practice, that could mean bringing relevant supplier and contract information to the point of decision, while directing unusual cases to an appropriate member of staff. The aim is not simply to accelerate separate steps, but to make the hand-offs between them more coherent and responsive.
Other findings from the Ardent Partners and Ivalua research underline why that ambition is difficult to realise. A PR Newswire announcement about the study says 87% of teams remain burdened by tactical work, while 67% are not ready to scale AI. Although 90% of respondents are exploring or piloting AI, only 23% have managed to scale it reliably. The announcement identifies data quality, structure and availability as a major obstacle for 59% of respondents.
The underlying data is often not joined up. In earlier analysis of the same research, CPO Rising reported that only 11% of respondents had a unified procurement data model across Source-to-Pay. A further 41% had integrated data but still needed manual reconciliation, while 40% described their information as fragmented. When supplier records, contracts, sourcing history, invoices and risk signals are held in separate systems, AI may lack the context needed to provide a dependable recommendation.
Governance is another condition for adoption, not an issue to address after deployment. CPO Rising’s coverage of the research found that 34% of organisations had no formal AI governance, while 96% regarded the risks of opaque, or “black-box”, AI as moderate or high. Nearly two-thirds of respondents said human oversight was essential. For systems that influence supplier decisions or execute workflows, staff need to understand what the technology has done, why it acted and how to intervene.
Procurement leaders’ stated priorities for 2026 reflect the scale of the transformation under way. In the study, 52% cited the use of new AI technologies, 46% digital transformation and 42% better internal collaboration. A third identified streamlining existing systems and processes. Taken together, those priorities suggest that technology investment needs to advance alongside improvements to processes, infrastructure and cooperation between teams.
But pursuing all of these objectives at once can make progress harder to sustain without a clear order of operations. Ardent Partners’ readiness framework, described by CPO Rising, distinguishes organisations with both strong AI ambitions and the foundations to act on them from those whose plans outpace their capabilities. Data, connected workflows, governance and workforce skills are among the factors that determine whether an organisation can move from pilots to reliable use at scale.
The research’s central implication is that procurement teams need to develop those foundations in parallel with their AI plans. A connected, governed operating model can give intelligent systems the information and boundaries they require, while preserving human judgement where it is needed. Without that groundwork, adding more tools risks automating isolated activities without changing how procurement works as a whole.
Source: Noah Wire Services
