How Procurement Teams Can Drive Cost Savings with Self-Service Analytics, via GenAI-Powered Chat

29 Oct, 2023 •

data insights for procurment

Budget constraints can be slashed when businesses utilize GenAI-Powered Chat. Responsibilities for procurement to efficiently manage spend is increasingly critical. Empowering the procurement team with the ability to gain immediate insights into their questions would help them to drive their objectives.  

The role of the procurement professional continues to expand and the promise of tools and technologies to remove day-to-day manual activities has not yet been fully realized. Additional barriers that block the procurement from becoming most effective are:  

  • Talent shortage. Finding and retaining staff is difficult.  
  • Onboarding cycle times can be high to train staff as companies experience higher turnover. 
  • High dependency on IT or other resources to create queries and run reports to answer complex questions.  
  • Translating data insights into actionable results: It is not always clear how all the permutations of possible sourcing options could play into different business outcomes  
  • Limited access to all data. Cloud procurement applications have some default out-of-the-box reporting, but this often does not have access to all the data needed to make better-informed decisions. Procurement professionals may have to access multiple systems, and they typically need to manually consolidate data. 
  • Data volumes are increasing: Identifying patterns in large data sets while determining what is relevant or not is not an easy task.  

To overcome these challenges, fast and iterative data analysis around urgent questions is required.  

The challenge: “in-the-moment, iterative analysis,”

To say the challenge is simply “ad hoc analysis” understates the problem. The ad hoc nature of analysis is also both “in-the-moment” and iterative. It is compounded by the traditional approach requiring both in-depth business intelligence (BI) and data skills. Often this means waiting for days for a report that may not have all the right information.  

Default BI procurement reports can give a reasonable starting point, and it is possible BI assets are sometimes developed by data analysis experts who understand how to access and navigate the data model behind the scenes. Often those efforts can solidly address the more common reporting requirements. But in many scenarios, the procurement staff needs to go much deeper and pivot to many other related questions – all very quickly and iteratively, and in-the-moment. 

Procurement staff prefer to hypothesize and analyze on-the-fly. However, the company cannot offer real-time access to the centralized BI team. Typically, procurement professionals do not have the deep BI skills needed to get answers within the minutes or seconds for a question “in-the-moment.” 

The solution: GenAI-Powered Chat interface that understands your data 

DvSum CADDI (“Conversational Analytics for Data Driven Insights”) securely combines OpenAI’s GPT-4 with a powerful underlying data infrastructure – including data catalog, data governance, and data quality. Therefore, any authorized user can query procurement data quickly, easily, and iteratively.   


With DvSum CADDI, any person, whether technical or not, can simply ask questions naturally and iteratively with their data from a web browser.  

Users are not constrained to understand how to ask for information in SQL or SQL-like query statement such as “Sort Spend by Category by Date, where Date = 4Q22- 1Q23”.  Instead, users can ask a question no differently than if they were talking to another person, for example, “Give me the spend by category from last year’s fourth quarter to this year’s first quarter.”  

CADDI understands the data landscape and automatically accesses multiple databases and systems to quickly provide an answer and the sources cited to confirm the reasoning behind the answers. Most importantly, CADDI provides ad hoc analysis that is iterative and in-the-moment, in context of:  

 1. No BI expertise required to use GenAI-Powered Chat.

  1. For the user asking the question, it means a BI report expert is not required to navigate a report with hundreds of dimensions and filters.  
  1. For the BI and analytics team, it means not having to be distracted from core work because the procurement staff can handle the query themselves. 

2. No expertise in the schemas or data landscape required 

Even when it is not clear which specific data elements are required to answer a question, CADDI can help. CADDI eliminates the need for users to understand how to speak like a data expert. Users no longer need to learn technical aspects about the data, for example: facts, dimensions, inline dimensions or why the same field name is in multiple fact tables. 

3. Advancing the conversation 

The key goal for procurement staff is to ask questions, form hypotheses, and then test those hypotheses and even pivot to related questions. With DvSum CADDI, professionals can get quick answers, iterate new questions against the results and pivot to other questions as required. All this means that the procurement staff can advance the conversation instead of being blocked until a question is answered by the BI team. Being able to achieve in-the-moment analysis iteratively saves time thus accelerating solutions.  

Example Procurement Use Cases 

a.    Supplier Consolidates 80% of Spend. 

With DvSum, users can drill into opportunities to savings, cost reduction, and cost avoidance.  Categories can be defined in a variety of ways, for example: temporary services, office furniture, catering, wood, packaging, corrugated, elevator maintenance, and so on.   

The following questions mimic a day-in-the-life of a procurement manager who is researching an opportunity to consolidate the supply base which can lead to adjusting the negotiation strategy. With DvSum, procurement professionals can quickly and easily ask a question with GenAI-Powered Chat and use the results to adjust or pivot the follow-on question. For example:  

  1. What is the list of categories managed by me? 
  1. For category X, how many suppliers make up 80% of the cost? 
  1. For category X, are all business units buying from the same supplier? 
  1. What are the top five suppliers across each business unit globally? 
  1. Are the top five suppliers currently under contract? 
  1. When do the contracts of the top five suppliers expire?  

b.   Price Variance 

This scenario, procurement professionals can quickly and easily determine outliers of spend for a given line item of purchase or purchases with respect to different dimensions such as time, geography, supplier and so on. Below are questions our users are asking: 

  1. What is the average price of copy machines? 
  1. What is the spend for copy machines by supplier and state over the past three years? 
  1. Which are the lowest price copiers over the past two quarters? 
  1. What are the most expensive copiers in the past two quarters? 


DvSum CADDI empowers the procurement professionals to become more autonomous and productive “in-the-moment” and drive cost savings and even reveal potential new business opportunities. Contact us today to see how DvSum CADDI can help you. 

Share this post:

You may also like