Author: Chad Symens

Back in 2011, Lowe’s announced Q2 financial results with anemic growth and flat same store sales.   To improve performance, CEO Robert Niblock and EVP merchandising Bob Gfeller are implementing Integrating Planning and Execution (IPE), which places an emphasis on putting the right product in the right store at the right quantity.

This new focus got our attention since we provide EDI 852 data analysis and reporting to Lowe’s vendors.  Putting the right product in the right store at the right quantity is exactly what vendors use EDI 852 to accomplish.  Localized merchandising is the right strategy for Lowe’s, but they may run into some challenges executing the strategy with vendor’s assistance.

Many vendors we work with have Lowe’s as a customer, as well as other ‘big box’ retailers.  Across the board, these vendors get EDI 852 from their big box retail customers and we help them analyze the data at a SKU/store level.  But many of these vendors choose not to use Lowe’s EDI 852.

Instead,  they opt to pull reports from LowesLink®.  LowesLink® is a fine system for pulling reports.  The problem is the reports offer a snapshot of performance, not an analysis system.  If the vendor does not have a database to store weekly SKU/store data, it is nearly impossible for them to analyze weekly sales effectively and efficiently enough to participate in localized merchandising strategies.  Lowe’s does have Vendor DART which offers analysis tools, but the most powerful tools are reserved for large vendors.

Weekly analysis of EDI 852 at a SKU/store level is the foundation of a successful program like Lowe’s IPE.  Vendors know their products best and there are simply too many products for Lowe’s staff to conduct weekly SKU/store level analysis.  For Lowe’s Integrating Planning and Execution (IPE) to be successful long term, they must get vendors actively using the EDI 852.

LowesLink® is a registered trademark of LF, LLC.

Many vendors have started to using EDI 852 data or retailer portal data for sales and retail merchandising, but so far, only a few are using EDI 852 data for forecasting of demand.  But the reality is, vendor inventory at stores is often too low to meet demand and the rates of out of stocks have been increasing. It’s not a big surprise that retailers are maintaining less inventory in stores in this retail environment; the cost of excess inventory is simply too high and open to buy dollars are at an all time low. But with proper forecasting of demand, a vendor can help the retailer to better manage inventory and avoid out of stocks. The great benefit of EDI 852 for merchandising planning is that it is store/SKU level data. Since a typical retailer is forecasting demand at a category and market level, the variability in the rate of sales among stores in a market can be large.

A more accurate model for forecasting of demand is to start at the store/SKU level, calculating an average rate of sale for the store/SKU and then based on the inventory on hand at that store, a weeks of supply. When the weeks of supply for a store/SKU has been calculated, the vendor can compare against the lead time to replenish the store and work to put a true demand driven supply chain in place. This model, while more intensive for the vendor to manage, usually creates a far different picture of inventory needs than simply market level min/max replenishment.

Wal-Mart’s Retail Link* web site is a rich tool, providing vendors with a well of data for analyzing sales and inventory. The problem for many vendors is that Retail Link provides data, but storing the data, calculating metrics and providing users with analytics and reports can be a time consuming chore. Just this week I have spoken to three Wal-Mart vendors, who have dedicated staff to the task of running Retail Link reports and turning them into reports for other users. From our research working with many Wal-Mart vendors, we have calculated data manipulation and reporting averages 18 hours per week for most Wal-Mart vendors. On an annual basis, that is nearly 1,000 hours of administrative time spent just preparing data for analysis and reporting. In addition to the administrative time, there is a data storage issue most vendors encounter. Since the data is UPC/store level data, most vendors do not have the ability to store each weekly data file at the store detail level. As a result, they lose critical detail, as well as the ability to calculate key metrics like inventory weeks of supply on hand, or comp year performance comparisons. If your organization is spending administrative time preparing Retail Link data for analysis, consider for a moment how much more efficient it would be to eliminate that work and instead focus your staff on data analysis. What stores are performing poorly, what stores are performing above average, do you have the correct amount of inventory at each store? These are all questions your analysts should be analyzing, but instead they are spending half of their week preparing data. It’s a simple matter of efficiency and resource assignment.

*Retail Link is a Wal-Mart software application and is not affiliated with Accelerated Analytics.

What is EDI 852?   EDI 852 is a standard data format used to transmit product activity data. Files are typically sent daily or weekly and will include sales activity by product, and for some retailers, inventory on-hand.  Activity is typically summarized at a distribution center level, unless store level data is deliberately selected. Some EDI 852 forms will also include pricing information, inventory on-hand but unavailable for sale, order point, order quantity, and order status. EDI 852 is provided as a text data file using special character sets to describe the coded data to the decoding software.

Over the past month or so, we have been rolling out new sales analysis for our Home Depot vendors on market performance.  Vendors to Home Depot using EDI 852 know buyers often want to receive sales analysis at a market level.  What we often find, however, is the analysis conducted does not take into consideration the number of stores in the Home Depot market.  To conduct a proper POS analysis of Home Depot markets you really have to take into account the number of stores in market.

For example Home Depot market 48 has 80 stores, while markets 138 and 160 have one store.  By summing the units sold from the EDI 852 to a market level and then dividing by the store count in that market, you can get an average sales per store for the market.  This metric is very useful in comparing the performance of the markets in a meaningful way.

In addition, you can use the SKU count in your plan-o-gram to further compare the performance of the stores at a market level.  You might also consider a store level sales analysis so you can rank your stores by sales volume A to E.  This lets you not only compare the store against the average sale in the Home Depot market, but also against its sales volume peers.  POS sales analysis by Home Depot market can yield some very actionable results, give it a try.

Many vendors use displays at retail customers to drive higher visibility and increased sales. A floor pallet or wing stack with product is a popular display program. These programs are often offered to the retailer at a discounted price and usually require the vendor to absorb both the discount and the cost of preparing the display. The objective is a short term boost in sales that hopefully creates long term repeat buyers. The question is, do displays work?

We have been studying this exact question for a number of clients who sell at Home Depot, Wal-Mart, and Lowe’s. Unfortunately, the impact of displays can be very difficult to analyze, for the following reasons.

· The SKU’s on a display are usually the exact SKU’s that are available on the shelf and as a result it is not possible to uniquely track a display sale vs a regular on shelf sale. You can look at the sales before the display shipped and after it shipped and calculate the change in sales, but it’s not an exact science.

· Shipping of products is increasingly going through a distribution center and not directly to the store. As a result, the exact store and delivery date of displays can be difficult to track without significant extra effort and coordination with the retailer.

· The retailer often exerts a significant amount of influence on store selection and the timing of when the display will be shipped. While this is understandable, it can be detrimental to the vendor and, unfortunately, sales.  Not all stores warrant a display, and for many products, the timing can have a large impact on sales.

· The execution of the display can be uneven across stores, since many store managers have a great deal of latitude on where the display will be located. In addition, store managers have latitude on the length of time a display will be available, and when the display is broken down and the inventory put into the regular shelf position.

What can be done to increase the measurability and effectiveness of displays?

· The most important action a vendor can take is to use a display SKU that is different from regular on shelf items. This is the ideal solution, but in reality this is very hard to do, because retailers do not like to have new SKU’s and most especially if the item is a short term sale.

· Work with the retailer to design display programs that can be accurately tested. There are several variables you can work to control: store quality, location, timing, display type. The key is to create test programs which have these variables controlled, so you can track their success. For example, you might choose 50 grade A (high sales) stores in two geographically similar markets and ship a display to them in the spring. Then ship the same display to the same stores in the fall. This provides the ability to understand if time of year impacts the rate of sales. You can then create a program that changes a different variable and test its impact. Operationally, this can be challenging because often the retailer must approve the display store list and timing, but working together, there is an opportunity to explain the benefits of testing and the cost realities born by the vendor that make the test essential.

Accelerated Analytics will be posting additional thoughts on displays in the future and welcomes your feedback and insights.

If you have a seasonal product, is it best to run promotions during the core selling season or during the down season? I’ve been analyzing this exact question for a customer this week and the data can make a case for both approaches. On the one hand, it seems like running a promotion like a rebate or floor display during the key selling season makes sense. In that case, the promotion is timed to when the consumer is likely to purchase your product. Or at least be thinking about it. On the other hand, if the consumer is likely to purchase your product during the key selling season anyhow, perhaps managing inventory and avoiding out of stock is the right strategy. If promotions are aimed at the non-core selling season when demand is historically low, doesn’t that provide the opportunity to increase sales? It’s a difficult question and I’d be interested in hearing Brand Managers experiences and insights.

Retail sales consultant Retail Metrics Inc. predicts that the 30 national retailers it tracks will this week post a 3.4% gain in sales for December at stores open for at least a year. That is atop a 3% gain for December 2009.

Overall, the National Retail Federation trade group forecasts that retail sales in November and December increased by 3.3% this year to $451 billion.

The positive gains made this holiday season appear to be sustainable. In December 2010 and already in the New Year our customers are engaging us to help with detailed custom data analysis for line review preparation. This is always a good sign of an improving economy. It means business leaders are ready to compete. They are looking for opportunities and making sure any weaknesses are explained, along with a plan for improvement. Underlying all of this is very careful point of sale data analysis, so that plans and recommendations are fact based. We have been carefully analyzing products across retailers, identifying trends in product sales, geographic trends, pricing variances, etc.  It’s amazing what EDI 852 can tell you if you have a good database with several years of history. Dig into your EDI 852 and get ahead of your competition.