Bad data remains a primary cause of out of stocks, invoice disputes, supply chain glitches and delays in getting new products to market, according to the first major report into the business benefits of product information management.
IT consultant The Yankee Group, which interviewed 350 executives at global suppliers, found that organisations that had started to clean up their master data had seen reductions of more than 25% in purchase order errors, invoice queries and errors, plus improvements in on-shelf availability.
They had also slashed the time spent communicating information to trading partners about new products, and changes to existing products such as new formulations or pack sizes and promotions.
There was also a growing awareness that projects such as RFID implementations and co-
managed inventory would not deliver the returns they promised if all they did was move bad data around more quickly, said Alex Schumacher of b2b e-commerce specialist GXS, which commissioned the research. “There is a growing recognition that RFID, CMI and CPFR must be underpinned by high-quality, accurate, standardised product data. People recognise that bad data is holding them back.”
Recent research by UK data cleansing experts Udex showed that one in four suppliers failed data quality checks due to problems such as incorrect case or product dimensions, only discovered when products didn’t fit in the shelf space, or pallets were refused at depots, said Schumacher.
His comments came as Cadbury Schweppes became the latest supplier to publish its product information in the Transora data pool. This will allow all of its retail customers to access its data in a consistent format. Currently, suppliers introducing a new product or making changes to an existing one have to communicate this to each retail customer separately in different formats - a lengthy, error-laden process.
Data management specialist Sterling Commerce said that if the vision of global data synchronisation were realised, products would all be described according to a global standard and suppliers would only have to publish data once, in one of the network of data pools.
Elaine Watson