A Path or a Ditch? Navigating Transformation Using Supply Chain Analytics

Organizations need to take advantage of analytics to transform procurement and supply chain functions.

Landscape 57a37eee5edf5

Organizational transformation is the new imperative. Embracing digital and technological innovation is necessary to drive the business forward. CPOs are under continuous pressure to respond from their organizations. Mounting pressure causes many procurement professionals to revert to old known ways of doing things, focusing solely on cost reduction. Beating up suppliers for cost savings becomes the default behavior.

Procurement organizations face a challenging task—transform from a race to the bottom on costs into an organization focused on total cost of procurement. Geopolitical, economic and supplier risk take on increasing importance for supply purchasing decisions. Those using cost-only metrics face uphill challenges trying to transform their organizations. Analytics can aid in delivering results CPOs need to show gains. They can provide the breathing room necessary for transformation. Analytics can jump-start the transformation process, allowing procurement to expand the conversation with stakeholders. This helps show procurement’s ability to be a value-added partner within their organizations.

New technology alone cannot transform the procurement function. Many purchase expensive new software or a software-as-a-service (SAAS) subscription to a new platform to solve their problems. Without careful consideration about technology's role in transformation, many risk purchases that may be abandoned or only marginally adopted.

New technology coupled with business and thought process changes are necessary to embrace a new paradigm. Three key advances within the technology landscape ease its role in transformation: the movement to cloud technology, in-memory databases and visualization technologies.

Traditionally, hosting in the cloud focuses on benefits for the IT organization. Of equal importance is the ability for fast deployment. Gone are the long hardware order lead times, inevitable backorders and installation delays. Cloud providers handle much of the infrastructure setup, with the customer responsible for integration to internal networks when necessary. Cloud applications are typically a plug-and-play investment, with trade-offs of customization made for speed of deployment. Purchasing must also consider how these purchases can affect finance and return on investment (ROI). Consider the shift from historical multi-year product purchases and ongoing licensing deals to pay-as-you-go services. Historical purchases were capitalized and amortized capital expenditure (CAPEX), whereas subscription-based, pay-as-you-go services are often operational expenditure (OPEX).

In-memory databases are providing significant advances in processing power. These are no longer a theoretical concept—they are already available. The power of in-memory databases is the ability to process large quantities of data quickly. In traditional systems, many are constrained to running reports with limited datasets due to performance degradation in transactional systems. No one wants the dreaded phone call from IT support telling them their reports are slowing down the system. Previous technology limitations significantly inhibited large-scale data analysis and required entirely separate reporting systems. Even these systems quickly reached problems with scalability and flexibility to adapt. Significant volumes of data are captured by enterprise resource planning (ERP) processes: How is this data enabling you to make different decisions? Are you leveraging your data to its fullest capabilities? What could the analysis of that data teach you about both your organization and your industry? In-memory databases provide advances that allow fast processing and eliminate the barriers that limited large-quantity record set analysis.

The final piece to the puzzle is visualization technology. Visualization technologies provide new ways to represent complex datasets in simpler terms. Large datasets can be analyzed and summarized making the data meaningful and actionable. Finding outliers or areas of focus is significantly faster and easier when data is summarized visually through charts and graphs. Visualization tells the story about your organization brought forward from your data. Once you understand the story, you can add your own plot twists to write the ending. This is particularly important for organizations trying to deliver new value internally, as it provides new communication mechanisms to expand the conversation for procurement. For example, consider the number of days in the purchase-to-pay cycle as shown in Figure 1 (above) on the vertical axis. This information is combined with the percentage of early payments made. The bubbles represent the amounts based on these criteria. For a company with net 60-day standard terms, quickly analysis emerges for areas needing focus. Visualization technologies allow for the summarization of large quantities of data in a format that is easy to understand to allow quick action for further analysis.

Companies embracing analytics as part of their procurement transformation face three challenges. First, analytics must be implemented, which may mean new tools or systems in some cases. There is a requisite learning curve to implementing new systems and processes. Second, the rise of Big Data and analytics is exposing a talent shortage and skills gap within business functions. It is difficult to hire this skillset, with many employers competing for a small number of people. For most organizations, development of existing staff is the fastest and likeliest path for success. New career paths and opportunities for staff abound for those with an analytical and data-oriented mindset. Existing employees also have the advantage of institutional knowledge of the organization. Finally, unless there is clear measurement and agreement on measurement of outcomes and responsibilities, organizations may struggle to accomplish continuous improvement or even maintenance of the status quo.

Organizations need to start taking advantage of analytics in order to transform the procurement and supply chain functions. It is difficult to take the time to completely establish a strategy and action plan, as the business must continue moving and delivering, while simultaneously transforming itself. Given these constraints, where can you begin on this journey to create procurement as a value-added partner?

First, you must understand what types of information you would like to get out of your data. Without a clear goal, any Big Data initiative may struggle to deliver. What would you measure, so that you can measure what you manage? Using a classic risk management approach, a framework of analytics opportunities can be built. The framework should be built by analyzing known-knowns, known-unknowns and unknown-unknowns. The first step on the journey is the most difficult—consider the framework below when establishing your analytics program.


These represent key performance indicators (KPIs) or internally created metrics. They use descriptive analytics to summarize and describe data. For those new to analytics, these are the most straightforward and easiest to understand. They form the organizational blocking and tackling, and keep the business running smoothly. Master data is a key area that can benefit from this type of reporting, particularly for those organizations centralizing procurement and standardizing procurement processes.

Known-knowns can identify things to consider for standardization. Building these analytics into operational processes helps ensure initiatives to correct business processes and data problems are sustained. As transformation occurs, new KPIs may be established to measure procurement’s changing role. A few examples of known-knowns include:

  • Payment types and number of payments. This can be used to support bank fee analysis for bank charges and initiatives to move to electronic payments.
  • Lost discounts. What amount of discounts are you losing by not paying within the discount window?
  • Invoice baseline date versus invoice receipt date. Are your systems tracking the time between these two dates to see if they are reasonable? How are you registering the dates for invoices in your system?
  • Payment terms usage. How many different types of payment terms am I using across my company? Should I have more options or less?
  • Overdue invoices. How many and what dollar total of invoices are getting stuck in my processes? This can form quantitative information to support invoice processing improvement initiatives.
  • Number of invoices processed per month. How many invoices are being processed through each of my channels? Many organizations have multiple channels of invoices, such as third-party networks, invoice processing and routing systems, employee travel and expense, electronic data interchange (EDI) invoices and manual invoices. What are all the ways you process invoices? This may also provide information to form the basis for standardization or investments in technology to allow standardization.
  • Open purchase contracts or purchase orders. How many are currently open? When were they opened, and what quantities or amounts are remaining? This type of analysis can catch processes when purchase orders may not be closed out properly.


These represent metrics that can be identified, but are difficult to quantify on a large scale. This requires utilizing a combination of descriptive and predictive analytics. Descriptive analytics capture the metrics, while predictive analytics can quantify what business process decisions may be costing your organization. Examples of known-unknowns include:

  • Weighted average days to pay for suppliers versus the payment terms negotiated. What are the differences at a geographic level? This type of analysis may expose many different potential problems, including regional differences, training problems and master data problems. If you are not paying according to terms, what is this costing you?
  • Payment terms overridden. If I have standardized payment terms, how often are these being overridden? Does this reveal a process/policy inconsistency? What is this costing in working capital?
  • Early supplier payments. Am I paying my suppliers before necessary due to my accounts payable payment process frequency? What is this costing my organization?
  • Projected cash needs. When are invoices due according to terms? Understanding cash needs helps better project operational funding requirements for treasury.
  • Spend analytics. Do I know how much business I am doing with my suppliers? Are there opportunities for me to move to category management or other value-based purchasing models? In absence of reporting on spend, it is difficult to identify consolidation opportunities.
  • Purchase price variance analysis. Where do I have purchase prices that differ from my standard costs? This affects gross margin either positively or negatively, depending upon the direction of the variance. How is this affecting my gross margin?


These are the most difficult analytics to implement, but provide new opportunities by analyzing patterns in your data. When you possess the data across your entire financial supply chain, the data can be joined across the financial supply chain, providing a comprehensive view that simply isn’t available in traditional transactional systems. Simulations and correlations between the results of steps in your processes can lead to new insights. These are patterns that may otherwise be difficult to see without the processing power of Big Data. Examples of unknown-unknowns include:

  • Single-source versus multi-source supplier risk. What types of risk are you bearing that would be better mitigated by a multi-source supplier strategy?
  • Optimal material pricing. Are there pricing efficiencies that could be accomplished by adjusting the reorder entry point or quantities?
  • Identification of risk associated with procurement of materials critical to sales process. Can you identify the amount of revenue at risk due to materials sourced by procurement? Consider the value add and different decisions that could be made regarding sourcing with this information.
  • Identification of supplier risk associated with race to bottom on supply costs. Would you put your suppliers out of business with your focus on cost?

There are no one-size-fits-all approaches for analytics. Analytics can help a company provide the appropriate support to jump-start organizational transformation. Using a framework of building analytics using knowns, known-unknowns and unknown-unknowns provides the flexibility for the organization to quickly utilize analytics as a path forward instead of veering off into the age-old ditch of pure cost reduction. These technologies are all available now—how can you use them, along with an analytics framework, to change the procurement conversation with your stakeholders?