Artificial intelligence (AI) is no longer a futuristic concept—it's a game-changing force that has the potential to revolutionize many aspects of business. While the application of AI in the procurement function has gotten a reasonable amount of attention, it is an area where we believe that smart deployment of AI can yield great impact. As we often see, the key issues holding executives back are not knowing where to start and how to ensure a return on investment.   

This series of short articles is meant to guide CPOs and procurement leaders as they explore and implement AI solutions in procurement to transform their function.   

Our experience, coupled with our own survey of procurement leaders conducted in 2024 indicates that AI investments in procurement and supply chain are still somewhat modest. While we do not want to add to the “bots are coming for your jobs” discussion, we do want to outline ways to drive productivity in procurement by enhancing efficiency, accuracy, and speed and by driving more impactful decision-making across spend categories in less time. 

Getting started is the hardest part   

Figuring out how and where to start can be overwhelming. There are so many questions ranging from defining the business case for investment in AI to managing data quality and integrity, to ensuring compliance with ethical standards and regulatory requirements, to managing adoption and change management, just to name a few.  Additionally, developing a framework for when to create a solution in-house (rather than using a partner or service provider) can be applied to the first and then subsequent use cases.  We’ll dive deeper into this topic in a future article as it is critical to develop in-house capabilities over time as the number of external partners must remain manageable and fit into the organization’s IT landscape.  

Putting these concerns aside for a moment, a very straightforward first step CPOs can take is to ask themselves this question: “What would I do if I had unlimited FTEs on my team?” This can help assess the magnitude of the transformation needed and provide a good initial vision of what is really in it for your organization in terms of analytics, required compliance checks, supplier market research, etc.   

Biggest pain points, biggest ROI  

Once you’ve identified a sizeable pain point facing the procurement organization, avoid the trap of a months-long due diligence. Instead get familiar with the options, looking for quick wins and learning from mistakes. Using the simple assessment approach described above, determine the source of the pain within the organization (e.g., contract checks, consistency of offer checks, in-depth analysis of bids). Next, research whether a high-impact use case for the problem exists and investigate its parameters for implementation while balancing benefits and costs. Focus on near-term results: pick use cases that have the strongest foundational pillars and can be executed quickly to show results.   

As expected, there are many use cases for AI in procurement and we plan to explore many in this series. Here, we will focus on several basic ones, where small AI pilots can be initiated with minimal investment and training and are easily scalable. These use cases can be launched quickly and serve as a great proof of concept to demonstrate the value of AI.    

This approach aligns with what we currently see offered in the marketplace by service providers. Our research (Figure 1) shows that the vast majority of AI offerings in procurement address the straightforward pain points of spend analysis, performance monitoring, and task automatization. A limited number of sophisticated solutions, such as fully autonomous negotiations (in use by Walmart since 2022) and AI-driven cost modeling, exist.  

 

 So, to get started, here are a few simple AI use cases for procurement organizations to consider:  

Writing and content creation: one of the most popular use cases for Generative AI is writing and content creation because it is efficient and generally results in high-quality output, does not require large-scale implementations and small pilots can quickly prove their worth: Examples include:   

  • Contracts  
  • Statement of Work (SOW)  documents 
  • Deliverables descriptions  
  • Requirements development  

Investigation: AI can quickly analyze documents for any terms out of compliance or dated terms and can pull contract terms into summaries for actionable insights   

Spend Categorization: make use of Generative AI (GenAI) to receive sorting logic and taxonomies for undefined spend categories to reduce the extent of a line-by-line comparison  

A number of tools that may already exist in your current tech stack support several of the functions described above:   

  • MS Copilot, which most companies have as part of their MS365 licenses, offers a basic solution for some critical tasks in procurement such as the ability to compare contracts/offers, analyze data, automate content creation, etc. Microsoft Power BI is also often included and can be used to analyze procurement data, generate insights, and create dashboards  
  • IBM Watson, often available through existing software contracts, can be used to help with the automation of some procurement tasks, contract analysis, and supplier risk assessments  
  • SAP Ariba and Oracle Procurement Cloud have AI capabilities for spend analysis, supplier management, and contract lifecycle management  
  • Salesforce Einstein can automate workflows, predict trends, and analyze procurement data  

 Thus, introducing AI solutions in procurement does not have to be expensive, but it should be aligned with corporate standards and policies governing the use of AI.  Limitations generally come from a lack of useful data and the additional complexity introduced into the company’s IT landscape can be a restricting factor. But even with limitations, the benefits that come from accurate contracts or the ability to identify new opportunities by screening contracts on a large scale outweigh the challenges.  

 Remember that it can be crucial to start small and scale up. Begin with pilot projects that address specific pain points and demonstrate clear value. This iterative approach allows for quick wins and builds momentum, making it easier to secure buy-in from stakeholders and justify further investments in AI. 

Four keys to success   

Integrate AI into daily operations – Experiment with AI, exploring different angles to solve problems and make AI part of workflows and business orders.  

Prioritize people, not just tech – AI’s impact relies on adoption across the organization, not just its implementation. Offer resources to help employees integrate AI into their day-to-day workflows.  

Design effective data management policies – Make crucial data accessible and focus on quality to train AI effectively. Invest upfront time for better accuracy and speed.  

Measure success – Track KPIs, success stories, and improvements in efficiency, performance, and spend management for future rollouts.  

What’s ahead in our series? 

As we delve deeper into this topic, we will explore more best practices and success stories to guide you through the AI transformation in procurement. In the next article, we will dive into some unique procurement use cases that can open the art of the possible to solve complex business problems.  

At AlixPartners, we are actively engaged in developing AI tools and strategies that support organizations in navigating the complexities of modern procurement. Our goal is to simplify the modern solutions landscape and guide you on when and how to incorporate them—while also identifying situations where AI may not be the right fit. 

Stay tuned for insights that will help you navigate the complexities of AI and harness its full potential to drive procurement excellence.