AI is moving faster than most organizations are ready for. That's the blunt warning behind the latest advice from Forbes: before you chase artificial intelligence in procurement, get your house in order.
The message comes as enterprises race to deploy AI tools across supply chains and purchasing departments, lured by promises of cost savings and efficiency. Yet many are skipping a foundational step: cleaning up their data and processes. Without that, AI projects in procurement are doomed to fail or, worse, deliver misleading insights.
The urging to slow down reflects a broader pattern observed across industries. Over the past two years, procurement technology spending has surged, with startups and legacy vendors alike touting AI-powered analytics, automated sourcing, and contract intelligence. But the reality is that most procurement organizations still struggle with fragmented data spread across legacy ERP systems, spreadsheets, and paper invoices. AI models trained on messy data produce messy outputs — and that's a risk few companies have fully grasped.
The Forbes article argues that organizations must first standardise their master data, map out end-to-end processes, and ensure governance policies are in place. Named examples of successful adoption — such as a major retailer that cleaned its supplier database before rolling out AI-driven category management — show that preparation is the difference between a pilot that scales and one that stalls. The piece also highlights that procurement leaders should invest in user training so teams can interpret AI recommendations critically.
Industry analysts echo this caution. A 2025 survey by Gartner found that nearly half of supply chain leaders cite data quality as the top barrier to AI adoption. Without a solid data foundation, even the most advanced large language models can generate incorrect vendor recommendations or miss contract compliance issues. The message is clear: AI is only as good as the data it consumes.
Looking ahead, the procurement AI market is expected to grow at over 25% annually through 2030. But the winners will not be the first movers — they will be the best prepared. Companies that take the time to fix their data hygiene and processes now will be the ones that reap the rewards later. The Forbes council member's advice serves as a timely reminder: don't let the hype outrun your readiness.
AI
→ Neutral
Before You Chase AI In Procurement, Get Your House In Order
AI is moving faster than most organizations are ready for.
Shaz Khan, Forbes Councils Member
Forbes
2 min read
4/10
Key Takeaways
- The Forbes article stresses that data standardization is the number one prerequisite for any procurement AI initiative.
- It warns that AI adoption without proper governance and clean master data leads to inaccurate, even harmful business insights.
- Procurement leaders are advised to start with a small, well-defined pilot on a clean dataset before scaling AI across the organization.
- Employee training in interpreting AI outputs is highlighted as a critical, often overlooked step in successful adoption.
- The piece notes that technology vendors frequently overpromise AI capabilities, obscuring the real work needed on the buyer's side.
Frequently Asked Questions
Data readiness ensures that the AI models trained on procurement data produce accurate and reliable insights. Without clean, standardized data, AI can generate incorrect vendor recommendations, miss contract compliance issues, and lead to poor purchasing decisions.
Companies should first standardize master data across all procurement systems, map out end-to-end processes, establish governance policies, and run a small pilot on a clean dataset. Employee training on interpreting AI outputs is also critical.
Common pitfalls include skipping data cleansing, underestimating the complexity of legacy systems, failing to involve procurement teams in AI design, and trusting vendor promises without verifying data readiness.
Teams can prepare by auditing existing data for duplicates, errors, and inconsistencies; implementing a standard taxonomy for suppliers and materials; and integrating data from disparate sources into a single source of truth.
Long-term benefits include higher success rates for AI projects, more accurate cost-saving predictions, better supplier risk management, and the ability to scale AI across the organization without repeated failures.
Projects often fail because the underlying data is fragmented or of poor quality. Even the most advanced AI cannot overcome dirty data, leading to unreliable outputs that undermine trust and adoption.
Topics
Original source
www.forbes.com
Discussion
Join the discussion
Sign in to post a comment or reply.
No comments yet. Be the first to share your thoughts!