You are here
Kum & Go LC doesn't employ an inventory specialist or loss prevention supervisor. Instead, these tasks fall on various departments at the West Des Moines, Iowa-based convenience store chain, which operates more than 430 c-stores.
So when the chain decided to automate its inventory and gain control of both out of stocks and overstocks, the implementation and use of Retalix's Demand AnalytiX (DAX) -- a browser-based, demand-driven replenishment solution -- fell into the lap of Kym Howe, senior vice president of IT. The DAX team uses the system, which offers demand forecasting, inventory, ordering and analytics, to uncover and report inventory issues at the store level using key performance indicators (KPI).
"The DAX team, which consists of two people who report to me in IT, looks for out-of-stock issues, uncovers the best and worst stores in terms of out of stocks, and finds the best way to attack the problem," Howe explained. "They can lay in any KPI to help them understand what is going on at a store, like what percent of orders are being changed by the manager."
The chain began using the system three years ago for computer-generated ordering, starting with cigarettes, and the benefits were immediate. They added more categories along the way and eventually rolled it out to all products supplied by its wholesaler in 2009.
"We reduced our overstock inventory by $2 million right away," said Howe. "We have always been about going over the top with customer service, so our managers would get into more trouble with an out of stock than too much inventory. They had no incentive to keep their inventory manageable."
To keep the payback coming, a score card was developed to measure the team's progress that projects a $3 million benefit to the business in 2010 through increasing efficiency, increased sales and a further reduction in inventory.
"Our out-of-stock rate is really good," Howe noted. "On average we are at 0.6 percent. The problem is we don't know what 'good' really is in the eyes of our customers. When we go into our stores, we still see out of stocks, but we generally only see them on slow moving items. Before having a system like this in place, we had no way of measuring out of stocks. Now, not only can we measure them, we can use the Analyzer tool to continually improve."
The DAX system also offers forecasting capabilities, with every order creating a new forecast. This enables the system to pick up on a trend quickly and factor it into the suggested order. But at first, managers pushed back.
"What makes DAX hard for store operations to understand is it also takes current inventory into account when generating the order. Our managers were used to a 'sell one, order one' mentality. DAX will take sales into account, but it first looks to see what inventory is on hand," said Howe. This puts pressure on the Kum & Go manager to always keep his or her inventory counts correct.
"Inventory accuracy is the key managers lose sight of," Tony Havlicek, planogram manager at Kum & Go explained. "Incorrect inventory will lead to overstocks/out of stocks if a store cannot accurately account for their inventory."
The DAX system helps manager do this through integration with the Autogas point-of-sale system, feeding real-time sales information to the corporate DAX server. The store handheld communicates directly to the DAX server wirelessly so inventory information on the handheld is always real-time, allowing managers to verify or correct inventory at any time. In addition, the DAX Smart Count module helped reduce labor time for item-level inventory management by telling the manager exactly what items he or she needs to count.
While the money saved on cigarette stock alone allowed the company to see a return-on-investment quickly, Howe admits it took her a while before she began using the more advanced business intelligence tools offered with the system. However, since incorporating the vendor's advanced DAX Analyzer tool into the system this past year, she realized what she was missing.
"DAX itself offers basic reporting, but this new Analyzer BI tool sits on top of the inventory data and is far beyond what the basic product offers," she explained. "They had told me about it, but it was pricey, and I wasn't sure if we needed it. I use Cognos for everything else, and I didn't want to have two BI systems, but we eventually decided to buy this rather than develop something ourselves."
The DAX team members are the primary users of the Analyzer tool, which is a memory resident tool based on QlikView platform. It allows them to quickly slice and dice data any way they want, and offers roughly 50 KPIs to choose from, Howe said. In the past the team could run reports and download them into Excel, because the system was never able to get all the KPIs wanted on one report. Also, some of the KPIs are only available with the Analyzer tool, she explained.
"I can run about 15 different reports by store or by item or by time fame, and the system is incredibly flexible and great for ad hoc reporting," Howe said. "It's also really fast to run. We have 432 stores and I'm throwing in five to 10 KPIs for a four-month time frame for all of them and it is available quickly."
The most important KPIs include out-of-stock numbers, inventory amounts, days of supply based on current inventory, average days of supply and out-of-stock revenue loss, which is the dollar amount lost for items that were out of stock based on their forecasted sales, she said.
"It finally allows me put a measurable cost to our out of stocks," she noted.
She plans to hit a return-on-investment on the new Analyzer tool this year with the scorecard, and each team member will work on a specific KPI.
"Now it's about tuning the system to make progress in the stores," she said. "The customizable Analyzer BI tool will allow the team to actually identify and make the improvements necessary."
Steven Kimmes, senior vice president of operations, added: "The compelling view is of the future. The ability to manage inventory at the item level is powerful, and migrating direct-store delivery vendors to computer generated ordering is an exciting prospect."