Procurement functions are increasingly being recognised for their potential to create value for the business at large. From cost containment and efficiency improvements to resilience, ESG reform, and relationship management, CPOs and their teams are balancing a seemingly ever-increasing number of plates. As a result, many procurement teams struggle to gain the necessary visibility and insights into the value chain needed to drive genuine transformation.
This is why data analytics are so essential to modern procurement. Data analytics can create the necessary granular insight into the procurement process and supplier ecosystem to achieve procurement’s growing array of goals. However, all types of data analytics are not created equal. Finding the right analytical tools and methods and applying them to the right problems is essential for procurement leaders.
Here are four types of procurement data analytics and the situations in which they are (and aren’t) applicable.
1. Descriptive Analytics
Descriptive analytics examine past procurement data to understand what happened in the past based on the data gathered at the time.
This is the simplest type of analytics to apply to procurement, but descriptive analytics are vital when analysing spending patterns, evaluating supplier performance, and incorporating purchasing behaviour into a broader procurement strategy.
Descriptive analytical tools are powered by data like historical purchases, requisitions, purchase orders, invoices, GRN, and tax receipts. They serve to help procurement teams understand historical data so that trends, patterns, and risks can be effectively identified.
2. Diagnostic Analytics
Diagnostic analysis is a slightly more complex process. Descriptive analysis focuses on the “what” of past procurement data. Diagnostic analytics go deeper to understand the “why”. They get at the root cause of a problem or situation that occurred in the past.
Diagnostic analytics are especially useful at investigating the causes of disruptions in the supply chain. By examining past and present data, diagnostic analytics tools can examine a supplier delay, for example, and provide insight into whether the event was an isolated incident or a recurring problem that might necessitate switching suppliers.
3. Predictive Analytics
As the name suggests, predictive analytics tools are used to predict future events based on past and present data. These tools, by necessity, tend to be much more sophisticated than descriptive and diagnostic analytics. They often employ artificial intelligence and machine learning to process vast amounts of data.
Used correctly, predictive analytics can identify future spikes (or drops) in demand based on past trends. This can allow procurement teams to adjust stock volumes accordingly and take other actions to avoid disruption. Predictive analytics are also capable of looking at broader data sets, including publicly available information like weather patterns, to identify potential disruptions to supply chains that could harm procurement.
4. Prescriptive Analytics
Lastly, prescriptive analytics harness the potential of cutting edge AI and machine learning. Prescriptive analytics can not only analyse data and make predictions, but recommend courses of action based on its findings.
As noted by Gartner, the “combination of predictive and prescriptive capabilities enables organisations to respond rapidly to changing requirements and constraints.”
There are a wide array of use cases that merge predictive forecasting and simulation with prescriptive approaches. These can include predicting infection risks during surgery and instituting rules to mitigate these risks. They can also extend to forecasting product orders to optimise responses to fluctuating supply chain demands, circumventing the use of potentially flawed historical data.
