Supplier Analytics: How Master Data Builds Trust

Large enterprises spend tremendous amount of resources building data analytics and operationalizing large volumes of supplier and spend data. However, analytics and dashboards are only valuable if the data going in them is trustworthy. Business end users are often faced with the question:

“Do I trust my supplier analytics?”

In most cases, business users consuming supplier data analytics don’t have a choice but to accept the data they are given. Even if they feel the data is a bit off, trying to figure out exactly what is wrong with the data (or if an individual is frustrated enough to attempt to fix it) requires a non-trivial amount of time and effort to triage issues with the IT or data team. This can be a huge liability, as various business decisions are made based on untrustworthy data.

In a previous article, Who Are Our Suppliers and How Do We Interact with Them?, we used a Mastered Supplier Dashboard to explore how an organization went on their DataOps journey to unify and master supplier data from disparate data sources. This resulted in an ongoing process where the organization knows exactly which distinct suppliers exist across their data systems, and the potential for duplicative interactions.

Using the Mastered Supplier Dashboard below, we continue to walk through why golden records for supplier data are critically important downstream for business users to trust their supplier analytics and dashboards.

Golden Records for Supplier Analytics

The Mastered Supplier Dashboard below tracks an organization’s data ops journey through mastering supplier data from disparate data silos of Ariba, SAP, and purchasing card systems.

The dashboard tracks data across 3 published dates:

• State of supplier data BEFORE mastering
• State of supplier data AFTER mastering
• More data sources and records are added, mastered, and tracked

See both the data impact (‘Mastered Data’ Screen), as well as the domain-specific business impact (‘Supplier’ Screen) from having curated, unified data.

A walkthrough of how mastered data impacted the organization’s end user view of suppliers is described in the section below. Actual data in this example is derived from USA Spend vendors.

Walkthrough of How an Organization’s Mastered Supplier Data Impacted Downstream Supplier Analytics

Use the dashboard to follow an organization’s journey to mastering supplier data and gaining accurate analytic insights.

The below timeline highlights how mastered supplier data can impact the downstream insights and every day operations for business users.

Problems Before Mastering

• The organization’s system shows over 54,000 distinct suppliers that end users interact with making up of $219 million spend
• The business users know this can’t be right, but can’t clearly explain why

Instead of trying to look for one-off examples and triage with the IT/data team, the business users know that the organization is going to use a machine learning approach to master their suppliers (at the site-level) and have results to iterate on within 1 week.

The business users also knows that the machine learning approach enables them to make direct impact to the mastering process. In addition, one-off feedback to data issues can be effectively collected directly from the dashboards through Tamr Steward (a Tableau extension not shown here) directly from the dashboard to resolve the issue.

Benefits After Mastering

• After mastering, the supplier data shows that the organization should be interacting with only about 41,000 distinct suppliers, and the other interactions are potential inefficiencies (that can be resolved for huge cost savings!)
• Their top 10 suppliers actually make up almost 10X more spend than they had previously been reporting
• The organization’s spend is also concentrated around a much smaller amount of suppliers than previously realized
• The view of the organization’s most important suppliers may completely change!

The example here demonstrates the impact of mastered suppliers on supplier concentration, which may lead to significant strategic and operational improvements.

Ongoing Data Unification

• The organization adds more data sources and the machine learning model from previous exercises continues to be applied seamlessly on new incoming data
• Business users get only distinct suppliers added to ongoing analytics for insights

Having a robust process in place to unify supplier data enables organizations to always have confident answers to their supplier questions without having to worry about data quality.

Unlock More Insights with AI-powered Golden Records for Supplier Data

This walkthrough highlights how golden records for supplier data can significantly impact an organization’s business view of their distinct suppliers, supplier interactions, and top suppliers.

In addition to these insights, agile mastering of distinct suppliers provides unified, domain-specific entity data that enables some astounding analytics. By grouping distinct suppliers together based on machine learning models (instead of rules), organizations can resolve (or leverage) discrepancies on contract terms, supplier distribution, supplier concentration, supplier risks, and more to achieve transformational outcomes.

Tamr has helped some of the world’s largest enterprises unify and master entity data. For one customer, Tamr was able to help consolidate unique suppliers across 75 ERP systems. This led to insights that immediately generated $80 million savings with an additional potential of $300 million.