Speaking at ARC-Retail’s first conference in September, Dr Phil Streatfield, Woolworth’s SC and IT director, told a simple tale of a hanger.
Dr Streatfield, who moved to Woolworth last year after pioneering innovative supply chain techniques at Entertainment UK, had been challenged over the poor out-of-stock position seen by a fellow director in his tour of stores. As a result, he picked a stock item at random and tried to discover both the theoretical and actual stock levels for a pack of humble plastic hangers.
Identifying actual levels in the 800 stores which stocked the line was comparatively simple – a telephone call to the relevant manager. Finding theoretical levels was rather different. As Streatfield admits, it took three intelligent people two days to find the information. They had to search seven disparate IT systems and found negative book stock levels, 33 stores out of stock and various ways of counting goods in transit.
”We found serious challenges around joined-up demand management,” Streatfield told delegates. ”We were trying to forecast demand without knowing what our lost sales were so the forecasts were inaccurate and systems had built-in inadequacies that compounded the errors.”
The Woolworth team set about trying to improve the situation by establishing a vast database of two billion records tracking SKU movements on a daily basis. They analysed sales against minimum credible display quantity (MCDQ) and how low availability affected sales. They looked at how, where and when goods were counted and how stock levels and availability were calculated and they crunched an awful lot of numbers. They haven’t found all the answers but the company is in the final stages of selecting systems to help solve the problem.
As well as highlighting holes in what appeared to be a joined-up supply chain, the year-long study also revealed a significant lack of understanding of the IT systems themselves.
Dr Streatfield asked delegates: ”Do you understand the logic underpinning you system? A logic that has been invented by a programmer or developer and is based on someone else’s view of the world. How much of the information these systems tell you do you take at face value?”
Goods-in-transit was a classic example: some systems transferred stock from distribution centre to store once it had left the warehouse. The product might linger in the back office for a while even though MCDQ on the shelf was not achieved. By the time goods were unpacked and MCDQ reached, a period of low sales had already affected the next replenishment cycle so demand forecasts were skewed, too little stock was delivered and out-of-stocks were inevitable.
Retailers have long been aware that on-shelf availability is not the same as store availability. It was an issue which Adrian Jones, W H Smith’s supply chain and merchandise director until the end of 2006, highlighted at The Retail Business Show earlier this year.
Jones told delegates how W H Smith now used independent assessors to monitor on-shelf availability in its stores. These figures are used in assessing store managers’ performance bonuses.
”Availability must be real to store managers. If they don’t understand the implications of poor availability, you’ll get nowhere”, Jones said.
Like W H Smith, Woolworth is now using teams of independent assessors to survey availability in store. This availability is then compared with theoretical levels to hone demand forecasting and the understanding of systems logic.
A joined-up supply chain is something many retailers strive for. Many may think they have achieved this Holy Grail, but that vexed question of availability remains. While some companies are using independent mystery shopper surveys, understanding on-shelf availability has also been a key driver in Marks & Spencer’s RFID project.
I once naïvely asked why, if M&S had an accurate system recording item sales in real time, did it need a nightly stock check to discover what was available? The reply hinted at shoplifters and human error but tactfully avoided the central issue that Dr Streatfield’s team uncovered. Too many supply chain systems are simply not joined-up and use variable logic to interpret data.
Penelope Ody is a regular columnist with SCS and is a retail market specialist
