The year was 1999, and things didn't look so bad for The Scotts Company. The maker of Miracle-Gro and other leading lawn and garden products had just acquired the Ortho brand from Monsanto, the latest in a flurry of acquisitions. In addition to dominating the U.S. market, Scotts had become number one in Europe. It was trumping the competition in nearly every category.

But all was not well. Executives knew the company had a seriously sick supply chain. Scotts was shipping product according to unreliable forecasts. It had no idea which items were actually needed, or where. To make up for its ignorance, safety stocks were unacceptably high. And communications with retailers were minimal. Says Mike Lukemire, senior vice president of global supply chain: "We were a terrible supplier."

To be fair, Scotts faced some unusual challenges. Vice president of operations Dan Paradiso describes the Marysville, Ohio-based company's supply chain as "violently seasonal." Nearly half of sales occur between April and June, as home and professional gardeners rush to buy product for the spring growing season. Scotts ships around $300m of product in April, and an equal amount in May. Levels dwindle to around $15m in October, and the company customarily loses money in the last quarter of the calendar year.
Making matters worse is the unpredictability of the weather. In 2002, heavy rains in the U.S. Northeast kept Scotts from having a decent sales weekend until the end of May, in a season that ends in July, says Sumantra Sengupta, senior vice president and chief information officer. Dara Mohsenian, an analyst with JPMorgan, blames bad weather for a drop in the company's North American consumer revenues in the quarter ending June 28, 2003 - a period that is expected to contribute 39 percent of full-year revenues in 2004.

Moreover, the company's supply chain is unusually complex. Scotts maintains around 800 distinct SKUs, varying widely in height, weight, moisture content and case configuration. For each one, it must craft a yearly entrance, replenishment and exit strategy within tight windows. The short season, combined with the weight and mass of fertilizer, dictates that most product be shipped direct to the store door. And small orders dominate, with up to 70 percent moving via less-than-truckload carriers, according to Paradiso.

But Scotts executives didn't point fingers at the elements, or other factors beyond their control. They knew the company's biggest problems were internal. Forecasting accuracy was a miserable 38 percent, with forecasts done only at the national level. Inventory was turning just 1.8 times a year. Customer fill rates were in the low 90-percent range. Essentially, says Lukemire, Scotts was filling up warehouses with fertilizer and letting retailers take over from there.

Much of the problem stemmed from multiple acquisitions, and the tangle of systems and processes that came with them. Retailers were having to deal with multiple sales reps, and the supply chain was highly fragmented. According to Sengupta, Scotts had 17 separate information systems in a decentralized organization. "There were redundancies everywhere."

Customer Comes First
The obvious answer was a supply-chain overhaul, focusing on the consolidation and centralization of key processes. Most of all, says Lukemire, the company was determined to become "customer-centric" - a term that is bandied about by virtually every CEO today. At Scotts, the words would take on real meaning.

Work had already begun on the manufacturing side. Lukemire, a veteran of Kraft and Nabisco, arrived in 1996 to address the problem. He oversaw a $150m investment in rebuilding the Marysville plant, applying lean manufacturing principles. Similar efforts were undertaken at the Ortho plant in Ft. Madison, Iowa, and at a durables facility on the West Coast. At all its plants, Scotts began practicing techniques of kaizen, the Japanese term meaning continuous improvement.

For enhancements to the larger supply chain, Scotts had little in the way of guidance. "The lawn and garden industry is about 10 years behind what other consumer-products industries are doing," says Lukemire. So Scotts found itself benchmarking against the efforts of leaders in other sectors, including consumer products giant Procter & Gamble, electronics manufacturer Flextronics, and some food suppliers. It even looked at companies such as Limited Distribution Services, which had set up an independent supply-chain management unit for its chain of apparel shops.

Business process innovation got underway with acquisition of R/3, the enterprise resource planning system from Walldorf, Germany-based SAP AG. It set the stage for creation of a single supply chain covering all Scotts products "from cradle to shelf," Lukemire says. The idea was to set up one place for the management of all sales, orders, shipping and customer service. At the same time, separate sales forces for Scotts, Miracle Gro and Ortho were combined into one.

The SAP system replaced a variety of homegrown and legacy systems, says Eric Domsky, the vendor's director of consumer products. Scotts' ultimate goal was to present one face to the retailer, which had previously coped with multiple orders, deliveries and invoices for the supplier's various product lines. Yet pulling together all of those systems was anything but easy. The implementation, says Domsky, "was like rewiring the house with the lights on."

Data accuracy was essential. Because it had maintained little interaction with customers, Scotts suffered from poor intelligence about consumer buying patterns. It took what was then a radical step: "We actually started having our supply-chain people talk to our customers," says Lukemire.

Business development teams were dispatched to the three biggest customers - Home Depot, Lowe's and Wal-Mart Stores - which today account for 70 percent of the company's business, according to Sengupta. Scotts' supply-chain personnel remain in place at the headquarters of each, working in tandem with sales staff. The program even got a catchy name: Collaborative Customer Centric, or C3.

The company also drew on SAP for a business intelligence platform, or SAP BW; a human resources package; electronic procurement; and warehouse management.

Tackling the Forecast
With key financial, sales and inventory systems in place, Scotts next tackled the job of supply-chain planning and optimization. It needed a tool for customer-based replenishment, fed by point-of-sale (POS) data.

Following an evaluation conducted by Cap Gemini Ernst & Young, Scotts chose Rockville, Md.-based Manugistics Group, Inc. for help in localizing and fine-tuning its forecasting model. Previously, Scotts had tolerated a large number of wasted moves, as it struggled to balance inventories among 18 mixing warehouses around the country. The Manugistics software specifies when to make product, where to deploy it, and how much safety stock to keep on hand at each location.

Scotts had been maintaining approximately three months of inventory for each SKU. Now it could assess proper stocking levels, based on statistical evaluations of every product. Warehouses could be redesigned to reflect real buying levels. No longer did Scotts have to resort to public storage in overflow situations, paying twice the rate of private facilities.

Demand forecasting and replenishment planning were the first modules to be installed, says Eric Symon, group vice president of Manugistics. Pilots with a select number of customers ran for six to nine months, followed by full rollout. As a result, Scotts was able to base its forecasts on real POS data, not what the manufacturing side predicted it would need to get through the season.

Manugistics also provided software that allowed Scotts to engage in Collaborative Planning, Forecasting and Replenishment (CPFR), the step-by-step process developed jointly by consumer-goods manufacturers and retailers to share sales data and make suppliers more responsive to actual conditions. Scotts brought together representatives of its planning, sales, information technology, supply-chain and human resources functions. Together they constructed a coherent business process for filling and replenishing retailers' orders.

More recently, Scotts implemented Manugistics' software for logistics management, as well as promotions planning and optimization. The goal of the latter is consumer-based replenishment, a practice that employs complex algorithms to determine the true impact of promotions, then uses the information to create better forecasts.

Integrating the various systems was not a problem, the principals say. "SAP from day one has been an extremely open solution," says Domsky. "There are no cases where we have 100 percent of the IT landscape." Symon says integration technology accounts for less than 10 percent of costs involved in such initiatives today. "The benefits of best-of-breed," he says, "far outweigh the cost of integration."

The Information Bridge
Helping Scotts to achieve external systems integration was Westboro, Mass.-based Ascential Software Corp. The vendor's DataStage software permits sales data profiling, quality assurance and transformation within a single platform. DataStage acts as a bridge between Scotts and its customers, accepting POS data from multiple retailers and loading it into the SAP BW business intelligence module. The consolidated information can then be used for planning, shipping and inventory management.

Previously, Scotts had no way of managing data quality or dealing with errors, says John Evans, director of enterprise application markets for Ascential. Anomalies didn't become visible until it was too late to correct them. Orders would have to be cancelled, then reloaded into SAP BW.

Integration was a major element in helping Scotts to get a handle on retailers' POS data. By the end of 2002, says Evans, it had reported a $99m reduction in inventory in the channel, compared with the prior year. At the same time, it realized a $158m increase in free cash flow generated from operations.

One of the driving forces behind the Scotts supply-chain project was a pressing need to increase inventory turns. That pathetic level of 1.8 stemmed from an inability to gauge true demand, causing Scotts to clog up the channel with safety stock. But when it came time to establish a target for improvement, executives aimed high. As of last fall, turns were in excess of five, according to Sengupta, who said at the time that "five and a half is best." That would mean about two months of supply in the channel.

Now, the company is saying it can do better. Lukemire revealed plans to reach 6.25 turns by 2006. Such a goal would be unachievable without the collaborative tools that Scotts has put into place with its retailer partners.

Of course, the company could boost turns simply by flushing product out of the channel, then praying that fewer safety stocks don't jeopardize its ability to supply customers. But Scotts is also determined to increase customer service, the other part of its two-pronged strategy launched some five years ago. And by all accounts, it is succeeding there as well.
"We've just announced record service levels with lower inventory," Lukemire says, citing "the power of information replacing inventory." Order fill rates now stand at approximately 98.5 percent. (The figure climbed to 93 percent in 2001, and 97.5 percent in 2002.) After a full year of running the Manugistics software, Lukemire is convinced that Scotts can get to 99.7 percent with six turns - at a lower cost of doing business.

Scotts has dramatically improved its utilization of resources all through the supply chain. Before, manufacturing plants were operating around 40 percent of the time they were scheduled to run. Now, the figure is closer to 85 percent, a level that Lukemire considers adequate, given the need to adjust manufacturing volumes according to the time of year. The number of mixing warehouses in the U.S. has been cut from 18 to 10, without hurting Scotts' ability to stock retailers' shelves. The creation of a core carrier program in Marysville picked up another $10m in savings this year, lowering the company's distribution cost by a full percentage point of sales, Lukemire says. Better planning also means it can take greater advantage of low-cost intermodal transportation, the use of which is up by 30 percent.

A Wealth of Savings
Lukemire estimates that Scotts has taken about $75m in costs out of its supply chain since 1998. He puts total cost savings related to the supply chain even higher, in the neighborhood of $125m. Meanwhile, company revenues have climbed steadily, from $1.7bn in the fiscal year ending in September of 2001, to $1.9bn in fiscal 2003. Net income has soared from $15.5m to $103.8m. Analysts estimate revenues for fiscal 2004 at $2.1bn.

Not all of the recent savings are a direct result of supply-chain initiatives. Lukemire credits the company's purchasing group for a large portion of its success. It is constantly grappling with the volatile nature of commodity pricing, particularly for urea, a key ingredient in fertilizer, and natural gas. JPMorgan's Mohsenian says rising urea prices could be a wild card in Scotts' financial performance this year. But so far, the company has managed to keep the situation under control, offsetting the impact of a 50-percent rise in urea prices. Lukemire says it has done a better job of wielding its considerable purchasing power to keep raw materials costs down.

The company's efforts have garnered approval from Wall Street. In a recent report, Mohsenian notes the "hiccups" caused by customer fulfillment issues related to the Ortho line, among other problems that damaged the relationships with retailers. But recent efforts "set the stage for a more efficient supply chain over the past couple of years with better retailer service," he says.

Today, says Mohsenian, "we view Scotts' competition as inferior, given what we see as Scotts' superior scale, marketing, innovation, distribution, and breadth of product offering." He expects the company to boost its already commanding 52-percent market share of the "do-it-yourself" lawn and garden sector in the U.S.

Joseph Altobello, an analyst with CIBC World Markets Corp., says Scotts maintains a strong advantage over its competition, bolstered by good fundamentals, investments in upgrading systems, and strong demographic trends. "It's got the best-known name brands in just about all categories," he says.

Retailers, too, have responded favorably. In 2002 and 2003, Scotts was named Wal-Mart's Supplier of the Year in the lawn and garden category. It was Home Depot's Partner of the Year in the same sector in 2003.

Much more remains on Scotts' to-do list. The company is still working to expand its CPFR effort into full-blown, consumer-based replenishment. Forecasts will be based on POS data for each item on a store-by-store basis, accounting for inventory constraints and other factors.

On the outbound side, Scotts is looking for new consolidation opportunities with its retailer partners. But there are also savings to be found upstream. The company will expand its use of purchasing consortiums from contract vendors to other suppliers. It means to grab control of inbound transportation as well, applying the same tools of network optimization and supply-chain modeling to that segment, Paradiso says.

High on the agenda is greater frequency of what Paradiso terms "untouched orders" - those that make it to the customer without issues related to backorders, credit approval, incorrect pricing or other glitches. Such orders made up 2 percent of the total four years ago; today they account for more than 60 percent. Says Paradiso: "We have basically engineered out our issues."

Lukemire says Scotts might even venture into selling its supply-chain expertise to other companies. The function is already a profit center internally, he says, but could be aimed as well at solving planning, purchasing and distribution issues for others.

The chief focus, however, will remain on Scott's relationships with its own retailers. The company is looking to fashion even tighter links between supply and demand, based on better communication of data with customers. And it means to take an even more direct hand in the distribution of product to stores. Says Lukemire: "I haven't found a buyer who, if you put data in front of them, will not allow you to manage their inventory."