Along with the current spate of profits warnings and “disappointing” results, we’ve all noticed how retail clearance sales seem to start earlier and earlier each year. Long ago and far away they were in mid-January and July, now mid-December and early June seem more normal. This year the “sales” seem to have been fairly continuous.
Clearing the remnants of old stock to make space for new season merchandise is as much a tradition in the non-food arena as the 5pm markdown of fresh produce that is clearly moving rapidly past its best, in the supermarket space.
Clearance sales and discounted promotions may make space on the shelves and boost turnover but they do very little to enhance profitable performance. Over the past few years “optimisation” has very much been the name of the game in retail supply chain execution. We have had them all: price optimisation, promotions optimisation, space optimisation and – of course – markdown optimisation. But, as more sceptical supply chain experts have also pointed out over the past few years, markdown optimisation is also an admission of failure.
In good times most retailers can live with such occasional failures. In bad times it is very different. Small wonder then that at this summer’s U Connect event in Dallas* forecasting was very much front of mind. “Forecasting was definitely on many companies’ agendas,” said John Radko, chief global technologist at GXS. “Discretionary spending is being badly hit in the current recession and all businesses operating in this sector need to match stocks to demand and improve performance.”
While for many retailers “forecasting” is more about gut feel and guesstimates there have been some notable exceptions: Entertainments UK, for example, under its then supply chain director Phil Streatfield, implemented a TXT solution that allowed it to forecast likely sales of new CD and DVD titles within hours of release. TXT’s first customers were in the fashion sector and it has become adept at helping companies predict sales of new lines based on historical performance: if you sell x in the first few hours then likely total market might be Nx during the life of the product, and so on.
Forecasting sales of staple or repeatable lines is rather different and this is where “flowcasting” has its supporters. This starts with forecasts for every product in each store rather than working on an RDC or national level. Once the store has made its sales forecast this can be correlated with existing stocks and required deliveries, which in turn are extrapolated to RDC requirements, over the time period under review, and ultimately to manufacturing production forecasts. Exponents claim that flowcasting can reduce out-of-stocks to just one to two per cent – compared with the more usual eight per cent-plus while total supply chain inventory can be cut by 30 per cent or more.
Reluctance to implement flowcasting techniques may, in part, be due to the fact that it requires significant collaboration between retailers and suppliers – and not many trading partners are, as yet, quite so eager to work that closely. Collaboration on forecasts involves trust and mutual confidence so neither partner is tempted to add in a little extra “safety stock” at each stage. Equally, flowcasting depends on long-term commitment with forecasts as much as a year out and good supply chain visibility to track products at every possible location from manufacturing plant to shelf.
Flowcasting depends on long-term commitment with forecasts as much
as a year out.Such long-term planning is clearly less easy during a downturn as it is difficult to predict just where consumer cutbacks will fall: equally, collaboration becomes even more important at such times. Those with memories of previous recessions will recall that retailers – especially the largest and most powerful – can be quite ruthless when it comes to cancelling orders and pressurising suppliers on price to maintain their own performance. In the days when there was a large and competitive supplier base to choose from they could probably get away with such behaviour.
In today’s global marketplace it is rather different – especially with those BRIC economies showing rather better growth opportunities than poor old Europe. If retailers want to keep their suppliers on-side then perhaps time spent on some collaborative and realistic demand forecasting sooner, rather than later when those orders need to be cancelled, might just prove to be a good investment.
*U Connect – now in its eighth year – is organised by GS1 US and VICS and is held each June.
Penelope Ody is a regular columnist for SCS and is a retail market specialist.