Vendor base rationalization is a critical aspect for many of the Global 2000 organizations. This is critical as duplication, inadequate or lack of information, lack of visibility etc. can impede an organization from achieving greater operational efficiencies. It is, therefore, essential to have a vendor master data that is de-duplicated, normalized & enriched and the vendor veracity is validated.

AutoNorm is an AI-based automated clustering and normalization software that is specifically designed for vendor master cleansing and classification. It provides repeatable cleansing to all "vendor based" spend data coming from disparate transactions systems. The Patent Pending Algorithm uses a training model specifically based on the existing vendor base of an organization, and can use Language Semantics to find similarity between differently named vendor.

• Fully automated software to find similar named suppliers and de-duplicate
• AI technology to help group suppliers logically within different geographies
• Normalization of all relevant vendor data
• Data enrichment adapters from D&B or Austin Tetra to check for diversity, credit rating for business partners and parent-child hierarchy
• Process bulk files at speeds exceeding 50,000 suppliers/hour
• Process historical suppliers database as well as new entries (on an on-going basis)
• Background check to identify blacklisted suppliers
• Cross-check with external agencies like GSA, Dept of Homeland Security, etc