AI in the Chain

Introduction

In 2025, procurement is undergoing a significant transformation, fueled by the growing maturity of artificial intelligence (AI) and a shift toward data-centric strategies. Rather than focusing solely on algorithms, leading companies are prioritizing the quality, structure, and integration of data to unlock AI’s full potential. According to Gartner’s recent insights, chief procurement officers (CPOs) are placing data at the center of their digital transformation strategies, using it as a competitive advantage to drive agility, cost savings, and supplier collaboration.

This article explores how adopting a data-centric AI approach is revolutionizing procurement by improving forecasting accuracy, enabling smarter sourcing decisions, and promoting ethical and sustainable practices.

Why Data-Centric AI Matters for Procurement

A data-centric AI strategy focuses on refining and contextualizing data inputs, ensuring they are relevant, standardized, and reflective of real-time conditions. Unlike a model-centric approach that emphasizes tuning algorithms, the data-centric model acknowledges that better data can lead to better insights and outcomes, even with standard algorithms.

Example: In a global FMCG company, AI-driven supplier risk scoring was underperforming due to inconsistent supplier data across geographies. By investing in centralized data standards and integrating supplier data feeds, the company improved predictive accuracy by 30% without changing the algorithm.

Applications of Data-Centric AI in Procurement

1. Smarter Supplier Selection and Risk Management

AI tools ingest structured supplier data—certifications, past performance, ESG scores—to generate reliable risk profiles and automate shortlisting. The more comprehensive and clean the data, the more accurate the analysis.

Example Prompt: “Evaluate suppliers by combining ESG ratings, on-time delivery rates, and risk events from the past two years. Rank top three suppliers for each region.”

2. Enhanced Category Management and Spend Analysis

Traditional spend analysis is often fragmented. With a data-centric AI strategy, companies integrate data from procurement platforms, contracts, invoices, and external benchmarks for deep visibility.

Hands-On Prompt Example:

“Analyze Q1-Q4 2024 procurement spend across categories. Identify areas of maverick spending and recommend corrective actions.”

Expected Output:

• IT Hardware Category: $1.2M spent, 15% maverick spend.
• Suggestion: Reinforce centralized supplier contracts.
• Office Supplies: $780K spent, compliance at 98%.
• Suggestion: Maintain preferred vendor agreements.

3. Predictive Demand and Inventory Planning

Accurate forecasting hinges on relevant and clean data. AI models trained with granular internal and external data—from order histories to market indices—enable better demand planning and sourcing decisions.

Example Prompt: “Forecast quarterly demand for office supplies across North America using 2023 order history and inflation indices.”

Expected Output:

• Q2 Forecast: 18% increase expected in print materials due to conference season.
• Q3 Forecast: Decline in demand for tech accessories as remote work stabilizes.

4. Improved Compliance and Sustainable Procurement

Data-centric AI helps ensure that procurement aligns with regulatory requirements and sustainability targets. Structured supplier data linked to carbon emissions, labor practices, and traceability allows for actionable decisions.

Example Prompt: “Identify suppliers with carbon emission levels above the EU threshold and recommend alternatives with verified green certifications.”

Expected Output:

• 3 suppliers identified exceeding limits.
• Recommended shift to 2 new suppliers with ISO 14001 certification.

5. Dynamic Pricing and Contract Optimization

Data-centric AI enables real-time market intelligence to optimize pricing strategies and renegotiate contracts proactively.

Example Prompt: “Analyze contract terms for packaging materials and compare with real-time commodity price indices. Suggest contract revisions if current rates exceed market average by 10%.”

Expected Output:

• Supplier X contract shows 12% premium over current market rates.
• Recommendation: Trigger renegotiation clause.

Challenges and Considerations

• Data Integration: Disparate systems and formats require robust data engineering.
• Governance and Ownership: Clearly defining who owns procurement data ensures accountability.
• Bias in Data: Unstructured or legacy data may embed historical biases that AI can inadvertently amplify.

The Role of Prompt Engineering in Procurement AI

To harness AI tools effectively, procurement professionals must develop prompt engineering capabilities. Well-structured prompts maximize relevance and reduce hallucination.

Good Prompt: “Compare the total cost of ownership (TCO) for laptops from Vendor A and Vendor B, including purchase price, support services, and 3-year energy consumption.”

Poor Prompt: “Which vendor has better laptops?”

By mastering prompt engineering, procurement teams can extract richer insights, automate repetitive tasks, and ensure alignment between AI output and business objectives.

Future Outlook for Data-Centric Procurement

As procurement teams mature their AI adoption, data-centricity will drive:

• Greater personalization in supplier collaboration.
• Proactive risk management through real-time insights.
• Integration of unstructured data (e.g., supplier news, ESG reports).
• Autonomous procurement workflows for tactical spend categories.
• Scalable ESG tracking and compliance monitoring.

Conclusion

The future of procurement lies in mastering data. A data-centric AI strategy doesn’t just enhance the power of algorithms—it ensures procurement decisions are contextual, timely, and aligned with strategic goals. As 2025 unfolds, procurement leaders must invest in data quality, integration, and prompt engineering to fully unlock AI’s value.

📢 How is your procurement team leveraging data to power AI strategies in 2025? Share your insights in the comments!

References

1. Gartner (2025). 2025 Trends for Chief Procurement Officers. https://www.gartner.com/en/articles/2025-trends-chief-procurement-officers
2. Gartner (2025). Why a Data-Centric Approach Is Key to Scaling AI. https://www.gartner.com/en/articles/data-centric-approach-to-ai