Rocket Science Retailing
Stores can increases sales, reduce markdowns by better managing inventory
Matching inventory to demand is a neglected part of managing for most U.S. retailers, reports the Harvard Business Review. But rich rewards are being realized by those (relatively few) companies that use rocket science retailing blending traditional forecasting systems with the power of information technology.
In Rome, for example, jewelry manufacturer Bulgari found that simply being out of stock on a single item was enough to reduce the stores revenue by 3.5%. Its now seeking ways to improve planning processes.
HBR surveyed 32 mostly cutting-edge companies, tracing their practices and processes and identifying four areas critical to the twin goals of reducing markdowns and having adequate inventory to serve customers. The areas are:
- Supply-chain speed.
- Inventory planning.
- Accurate data.
From 1971 to 1995, department store markdowns grew from 8% to 33% of store sales in the U.S. This includes promotional and forced markdowns.
At most retailers, forecasting depends on the instinct of a few individuals, not science. It should include both. Retailers can significantly improve forecast accuracy simply by updating predictions with early sales information something most dont do, according to HBR. One retailer, for example, ordered garments 11 months before the product was even available.
Japanese fashion retailer World Co., however, analyzes each product at predetermined periods in the sales cycle and reorders items that may end up in short supply. Its gross margin on inventory investments: 300%.
Dallas-based Comp-USA, a computer retailer, found that even one or two days of early sales information can be useful in predicting sales. Buyers watch the sales of new products and update forecasts based on these projections. They expedite orders for items selling better than expected and, when possible, decline items that have not been shipped.
Only nine of the 32 retailers studied said they analyzed the accuracy of their forecasts, a step HBR identified as fundamental to improving forecast accuracy. More important is being able to react early in the case where forecasts are wrong.
World Co. predicts accuracy by displaying new products in a room, just as they would be in a store. About 30 employees, chosen to represent the target audience, estimate the likely success of each one. Products that generate more disagreement among these employees have been found to have less-accurate forecasts.
Heeding test results can be a problem also. Seventy-eight percent of the retailers studied test new products, but most buyers said results showing that products will be unsuccessful are often ignored. Instead, merchants blame the weather or some other factor and plow ahead.
Links in the Chain
Slow supply chains are an issue also. Retailers must commit to ordering before any sales information is available. As one merchant reported: We do pay attention to our tests. The problem is we already own the product; the test merely reveals it will be a dog once it gets to the stores.
World Co. attacks the problem by working closely with its vendors before even placing an order. It stores fabrics for which forecasts tend to be more accurate than those for finished goods and reserves production capacity in anticipation of demand. If a product is a hit, World Co. can manufacture and deliver additional units to stores in two weeks.
World Co. also empowers employees in product design, merchandising, operations and retail to make many decisions, avoiding bureaucratic delays.
Planning to Succeed
Inventory planning involves deciding when and how much to order or produce of raw materials, components and finished goods. Its different from forecasting because a planner might want more or less than predicted demand. At the household level, you might buy certain items toilet paper, for instance well ahead of demand while you order items such as bread and milk weekly.
One of the most glaring shortcomings for many retailers is they dont track when theyre out of stock and the resulting lost sales. Only 13 of the 32 companies HBR surveyed track stockouts; of these, only 11 estimate lost sales.
Tracking stockouts could help retailers set optimal inventory levels and appreciate the value in improving supply-chain responsiveness.
Its difficult to know how much sold-out product would have sold had more been available, but a two-step procedure can help. First, calculate the underlying demand based on sales when it was in stock. Second, combine the estimated demand rate with the duration of the stockout to estimate lost sales.
According to HBR, this technique can approximate lost sales to within 2% at the store level or with higher accuracy at the supply-chain level or for a category of products. One retailer found sales could be improved by about 10% simply by increasing inventory.
Accuracy in Data Collection
Contrary to popular opinion, most retailers have considerable difficulty capturing and maintaining accurate and accessible sales information. In the apparel industry, returns are often handled poorly. In the grocery industry, clerks may not know the difference between a medium tomato and such specialty types as organic or vine-ripe. So everything thats soft and red gets checked out as a medium tomato.
Similarly, distribution centers ship the wrong sizes or fail to record when vendors change the dimensions of their packaging units.
Is it any wonder a recent survey found a third of consumers who enter a clothing store intending to buy something leave with nothing because they couldnt find their size in stock?
Office-supplies retailer Staples responds with a zero-balance walk. An employee walks through the store each day looking for SKUs that are out of stock. Other employees then verify the surge in demand, data error or other factor that caused the sellout.
by Mark E. Dixon