Faced with vast amounts of data from multiple sources, supply-chain leaders are increasingly turning to management strategies that ensure quality, availability, usability and consistency throughout the end-to-end data lifecycle. 

These three core components will help them drive lasting success.

People

To standardize, align and improve efficiency and quality in data management, organizations are establishing roles like supply-chain master data managers and analysts to manage data aligned with organizational strategy. Important skill sets include project management, statistics, model building and data management. These roles improve security, access, collaboration and decision making. 

Successful organizations create centralized teams that consolidate data responsibilities among select individuals, with additional decentralized data owners in business units. With defined roles, organizations can ensure understanding of data’s impacts along the supply chain. 

Process

Data quality makes all the difference, and fewer people with manual access means fewer mistakes.

A key step to ensure quality and integrity is establishing “data dictionaries” for master data related to materials, customers and suppliers. These dictionaries provide information about business definitions, rules and constraints — plus guidance for data fields including data type, length, valid values, default values and relationships to other fields. 

From there, developing organizational playbooks can ensure a single source of truth for master-data documentation, including standard operating procedures, service level agreements, points of contact and data dictionaries. End users and stakeholders have full access to the playbooks. 

Technology

Robust data governance processes and machine learning technologies can reposition supply-chain data as a vital corporate asset. With intelligent auto-population, organizations can reduce time spent on manual data entry. 

In one example, a supply-chain data team created a chatbot using machine learning algorithms to interpret queries and surface answers. The chatbot responds to repetitive questions and reduces employee time spent searching for information.

Reference templates — or decision trees based on business rules and knowledge — identify similar materials that are referenced to auto-populate attributes for new materials. A supervised machine learning model auto-populates data by combining criteria from common material characteristics.
Combining reference templates and machine learning expedites cycle time, improves data quality and reduces the risk of knowledge loss.

Outlook 

• Strong data management helps organizations achieve competitive advantages such as:
• faster cycle times,
• higher data quality,
• lower inventory carrying costs,
• higher inventory turns,
• more effective risk and audit management and reporting, and
• better supplier and customer relationships.

Focusing on people, process and technology for effective supply-chain data management improves the accuracy of supply-chain planning and execution and strengthens the underpinnings for digital transformation.

Marisa Brown is senior principal research lead, Supply Chain Management, at APQC.