Procurement leaders are no longer asking whether artificial intelligence will matter. CPO Rising revisits Ardent Partners: A CPO’s Guide to AI Readiness, a discussion with Andrew Bartolini, founder and chief research officer at Ardent Partners, and Nick Heinzmann, head of research at Zip. The focus is how fast teams can use AI under inflation, tariff swings, and savings pressure without chasing hype.
Three Layers of AI in Buying Tools
Machine learning and natural language processing have lived inside procurement software for years, often out of sight. What has changed is the stack: deep learning, generative models, and agentic systems that interpret data, reason, orchestrate steps, and in some cases run tasks end to end. Generative AI drafts RFPs, contracts, supplier messages, and analysis narratives. Agentic AI chains models to prepare intake, move workflows, or guide stakeholders through complex steps. Understanding the difference helps leaders test vendor claims against real capability.
Data Before Models
Successful programs start with data orchestration, not model selection. Procurement needs supplier records, contracts, invoices, and purchase orders alongside external ESG scores, risk signals, financial data, and sanctions lists, plus stakeholder intake and onboarding inputs in one governed layer. When data stays split across CLM, ERP, ITSM, finance, and risk tools, agents cannot produce consistent results. CPO Rising notes that IT partnership on governance, access, quality standards, and refresh cycles pays off even for teams early in the journey.
Where to Start
Ardent Partners and Zip advise targeting processes with high volume, manual effort, and stakeholder friction. AI-powered intake can turn static forms and chat noise into guided, policy-aware conversations. Generative RFP and bid analysis speeds drafting from templates and history while scoring responses and flagging risk. Contract review surfaces auto-renewals, SLAs, liabilities, and compliance gaps in long PDFs. Agentic execution, still emerging, can update supplier profiles, chase missing information, and coordinate approvals so teams focus on judgment rather than admin.
People and Proof
Adoption fails when training and communication lag. CPO Rising emphasizes practical literacy: how outputs are produced, how to validate them, and how routine work shifts toward strategy, supplier relationships, and commercial calls. Impact should be measured, cycle time, productivity, stakeholder satisfaction, output quality, manual effort removed, usage, and business value, not deployment alone. Research involving more than 300 CPOs, as summarized by CPO Rising, shows 23% expect AI to be transformational within two to three years, 50% expect significant impact, and nearly 90% anticipate using AI by year-end. Because AI often arrives embedded in platforms teams already use, unprepared organizations risk slow uptake and inconsistent practice.
